Generate music with lyrics from text using ACE-Step
Modify a portion of provided audio with lyrics and/or style using ACE-Step
Extend the beginning or end of provided audio with lyrics and/or style using ACE-Step
Generate music from a lyrics and example audio using ACE-Step
Generate music from a simple prompt using ACE-Step
Explore interactive worlds in real time with movement and camera controls.
Alibaba's #1-ranked Happy Horse 1.0 — generate 1080p video with synchronized native audio and multilingual lip-sync from text prompts or images.
Generate 1080p video with synchronized native audio from a text prompt and references. Aspect ratios: 16:9, 9:16, 1:1, 4:3, 3:4. Duration: 3–15s.
Generate 1080p video with synchronized native audio from a text prompt. Aspect ratios: 16:9, 9:16, 1:1, 4:3, 3:4. Duration: 3–15s.
Happy Horse 1.1 is Alibaba's #1-ranked video model. This image-to-video endpoint animates a still image into 1080p video with synchronized native audio and multilingual lip-sync
Happy Horse 1.1 is Alibaba's #1-ranked video model. This reference-to-video endpoint turns up to 9 reference images into 1080p video with synchronized native audio and multilingual lip-sync for consistent characters.
Happy Horse 1.1 is Alibaba's #1-ranked video model. This text-to-video endpoint generates 1080p video with synchronized native audio and multilingual lip-sync from a text prompt alone.
HappyHorse video editing supports advanced video editing through natural language instructions. It allows for local or global editing of video elements using up to 5 reference images.
Generate natural multilingual speech from text with fast voice and language control using Qwen Audio 3.0 TTS Flash.
Edits images from one to three reference images and a natural-language instruction, preserving key details such as facial features and identity while applying the requested changes
Generates images from a text prompt at resolutions up to 2048×2048, with automatic prompt rewriting and prompt-guided resolution selection, building on Qwen's strength in complex text rendering and precise prompt adherence
Wan 3.0 is the latest generation AI video model, delivering enhanced motion smoothness, superior scene fidelity, and greater visual coherence.
Wan 3.0 Prime Image-to-Video turns still images into dynamic, cinematic sequences with rapid turnaround, natural motion, and excellent visual continuity. It preserves the identity, composition, and atmosphere of the source image while introducing expressive movement, camera dynamics, and richly detailed animation.
Wan 3.0 Prime Reference-to-Video combines reference images, videos, and audio into a unified video with fast generation and strong multimodal coherence. It follows character identity, visual style, movement, and sound cues across references to create controlled, consistent, and production-ready results.
Wan 3.0 Prime Text-to-Video transforms written prompts into polished videos with accelerated generation, fluid motion, strong scene fidelity, and coherent visual storytelling. Built for fast creative iteration, it brings complex ideas to life while preserving visual detail and cinematic consistency throughout each shot.
Wan 3.0 is the latest generation AI video model, delivering enhanced motion smoothness, superior scene fidelity, and greater visual coherence.
Wan 3.0 is the latest generation AI video model, delivering enhanced motion smoothness, superior scene fidelity, and greater visual coherence.
Generate professional-quality voiceovers in seconds with Async TTS Pro model text-based control over pauses, emphasis, and timing. Voice ids can be found at https://async.com/developer/voice-library
A audio understanding model to analyze audio content and answer questions about what's happening in the audio based on user prompts.
Upscale your images with AuraSR.
A fast and high quality model for image background removal.
A model for high quality and smooth background removal for videos.
bilateral reference framework (BiRefNet) for high-resolution dichotomous image segmentation (DIS)
bilateral reference framework (BiRefNet) for high-resolution dichotomous image segmentation (DIS)
Video background removal version of bilateral reference framework (BiRefNet) for high-resolution dichotomous image segmentation (DIS)
FLUX.3 is Black Forest Labs' frontier audio/video model. Re-render a previously generated draft at full quality — same seed, same motion, no re-planning.
FLUX 3 Image Edit from Black Forest Labs makes precise local edits without changing the rest of the image, and combines up to 10 references into one balanced, well-composed result.
FLUX.3 Edit Video [FAST] is Black Forest Labs' frontier video model. This endpoint edits an existing video from natural-language instructions, applying targeted changes while preserving the rest of the scene.
FLUX 3 is Black Forest Labs' frontier video model. This endpoint continues an existing clip beyond its final frame, generating additional footage that stays consistent with the original motion and scene.
FLUX.3 is Black Forest Labs' frontier audio/video model. Generate fast, low-cost draft previews that continue an existing clip, with a reusable draft cache for full-quality enhancement.
FLUX 3 is Black Forest Labs' frontier video model. This endpoint generates the video between a defined start and end frame, interpolating a smooth, coherent transition from the first image to the last.
FLUX.3 is Black Forest Labs' frontier audio/video model. Generate fast, low-cost draft previews between a start and an end frame, with a reusable draft cache for full-quality enhancement.
FLUX 3 is Black Forest Labs' frontier video model. This endpoint animates a single still image into video, extending one frame into coherent, natural motion.
FLUX.3 is Black Forest Labs' frontier audio/video model. Generate fast, low-cost draft previews that animate a still image, with a reusable draft cache for full-quality enhancement.
FLUX 3 is Black Forest Labs' frontier video model. This endpoint builds video from a sequence of keyframes, generating the motion between each anchor point for precise control over how a shot progresses.
FLUX.3 is Black Forest Labs' frontier audio/video model. Generate fast, low-cost draft previews pinned to your keyframe images, with a reusable draft cache for full-quality enhancement.
Edit images with FLUX 3 Image Klein base and up to 3 custom LoRAs.
Generate images with FLUX 3 Image Klein base and up to 3 custom LoRAs.
Fine-tune FLUX 3 Image Klein with your own images to teach it a subject, a product or a style.
Fine-tune FLUX 3 Image Klein on before/after image pairs to teach it a custom edit, with up to 10 input images per pair.
Edits and composes images from text instructions and up to ten references with FLUX.3 Klein, with optional bounding-box controls for object placement and transformation.
Edit images fast with distilled FLUX 3 Image Klein and up to 3 custom LoRAs.
Generates images from text prompts with FLUX.3 Klein, with configurable image size, reproducible seeds, prompt expansion and optional bounding-box layout control.
Generate images fast with distilled FLUX 3 Image Klein and up to 3 custom LoRAs.
fal’s FLUX 3 Max is a post-trained variant of FLUX 3 Video, tuned for stronger prompt adherence and better visual quality while optimized for lower latency.
fal’s FLUX 3 Max is a post-trained variant of FLUX 3 Video, tuned for stronger prompt adherence and better visual quality while optimized for lower latency.
fal’s FLUX 3 Max is a post-trained variant of FLUX 3 Video, tuned for stronger prompt adherence and better visual quality while optimized for lower latency.
FLUX 3 Image is Black Forest Labs' newest image model. Generate detailed, well-composed images in native 2K and 4K with precise layout control and improved text rendering.
FLUX 3 is Black Forest Labs' frontier video model. This endpoint generates video directly from a text prompt, translating a written description into motion, composition, and scene.
FLUX.3 is Black Forest Labs' frontier audio/video model. Generate fast, low-cost draft previews from a text prompt, with a reusable draft cache for full-quality enhancement.
Upscale videos to 1080p, 2K, or 4K via API. FLUX 3 powered super-resolution with a precise mode and a creative detail-enhancement mode.
Text To Image Model using Boogu-Image
Image To Image Model using Boogu-Image
Turn any flat ad image into fully editable layers —background, product and logo cutouts, live text with typography, and vector shapes. Commercial-safe, structured JSON output
Bria RMBG 2.0 enables seamless removal of backgrounds from images, ideal for professional editing tasks. Trained exclusively on licensed data for safe and risk-free commercial use. Model weights for commercial use are available here: https://share-eu1.hsforms.com/2GLpEVQqJTI2Lj7AMYwgfIwf4e04
Bria Background Replace allows for efficient swapping of backgrounds in images via text prompts or reference image, delivering realistic and polished results. Trained exclusively on licensed data for safe and risk-free commercial use
A high-fidelity capability for erasing unwanted objects, people, or visual elements from videos while maintaining aesthetic quality and temporal consistency.
A high-fidelity capability for erasing unwanted objects, people, or visual elements from videos while maintaining aesthetic quality and temporal consistency.
A high-fidelity capability for erasing unwanted objects, people, or visual elements from videos while maintaining aesthetic quality and temporal consistency
Seamlessly embed products into any scene with pixel-perfect control, automatic perspective, and natural lighting. Trained on licensed data - risk-free for advertising and eCommerce production.
Removes masked objects. For two erase modes: https://fal.ai/models/bria/fibo-edit-1.5/eraser
Bria Expand expands images beyond their borders in high quality. Trained exclusively on licensed data for safe and risk-free commercial use. Access the model's source code and weights: https://bria.ai/contact-us
Bria Extract Object uses text prompts to isolate a selected object from an image and return it as an RGBA PNG with a transparent background. Ideal for product, ecommerce, advertising, and creative editing workflows. Bria's Extract Object API leads in product shot extraction, outperforming SAM 3.1 where it counts most for commercial use.
Fibo 2.0 Edit makes targeted, instruction-based edits from up to four images. Background swaps, product restaging, keeping the product true to the original. Trained only on licensed data, with clear commercial rights.
Fibo 2.0 Generate is Bria's text-to-image model, strongest on reasoning and knowledge-heavy prompts the ones that need real understanding, not just a style match. Trained only on licensed data, with clear commercial rights.
A preview to the next level of control of Text-to-Image models.
Commercially safe, multi-reference image editing model. Follows natural language instructions alone or with up to 4 reference images, purpose-built for complex object and character combinations, virtual try-on, background replacement, style transfer, and more.
Bria Expand extends an image beyond its borders to any of 9 aspect ratios or a custom canvas up to 25 megapixels. Built on FIBO-Edit-1.5, with optional prompt guidance for the new area and the original pixels kept at full resolution.
Bria Product Holding edits a person photo to show the subject holding or carrying a product, using one to three product reference images and optional text instructions. Built on FIBO-Edit-1.5, it preserves the source aspect ratio by default and supports a selectable output aspect ratio.
Bria Virtual Try-On edits a person photo to show the subject wearing garments or accessories from one to three reference images, guided by optional text instructions. Built on FIBO-Edit-1.5, it supports multi-garment changes and preserves the source aspect ratio by default.
Precisely insert new objects into images with structured spatial commands. Context-aware, high-quality editing with seamless blending. Trained on licensed data for risk-free commercial and brand-safe use.
image composition model. Combine and blend multiple image parts into complex compositions through natural language and sequential editing.
Image colorization and color-grading model. Bring color to black-and-white photos or apply curated color treatments using simple style-based commands.
High-fidelity image editing model with state-of-the-art controllability. Combines JSON + Mask + Image for precise, fine-grained edits ideal for production and enterprise workflows. Trained on licensed data - safe for commercial use.
Structured Instructions Generation endpoint for Fibo Edit, Bria's newest editing model.
Remove unwanted objects from images with a text prompt - fast, precise editing that seamlessly blends results. Built for production scale and trained on licensed data for safe commercial use.
Precise, controllable photo re-lighting with structured text inputs. Apply natural lighting styles, soften harsh shadows, and transform scene illumination - production-ready and trained exclusively on licensed data.
Replace any object in an image using plain language with fine-grained, precise edits and strong prompt adherence. Trained on licensed data for risk-free commercial and brand-safe use.
Transform the season or weather of an image - summer to winter, sunny to rainy - with realistic atmosphere and lighting. Trained exclusively on licensed data for risk-free commercial use.
Photo restoration model that automatically denoises, deblurs, and enhances old or damaged photos - removes imperfections while preserving original character.
Production-grade style transfer that maps photos to distinct artistic styles using curated, brand-safe presets. Trained exclusively on licensed data for risk-free commercial use.
Precisely rewrite text inside images while preserving typography, fonts, and layout. High-quality, brand-safe edits trained exclusively on licensed data for safe commercial use.
Convert line drawings and sketches into photorealistic, fully colored images with preserved structure. Trained exclusively on licensed data for safe commercial and design use.
Text-to-image model with high-fidelity outputs, accurate typography, and style preset, strong in photorealism, textures, and beyond. JSON-structured prompts give enterprise and agentic workflows production-ready control. Trained on licensed data.
SOTA open-source text-to-image model delivering high-fidelity outputs with accurate typography. JSON-structured prompts provide production-ready controllability for enterprise and agentic workflows. Trained exclusively on licensed data.
Structured Prompt Generation endpoint for Fibo, Bria's SOTA Open source model.
Fast, low-latency text-to-image model with high-quality output and full JSON-structured controllability. Open-source, trained on licensed data, and optimized for production-scale generation.
Convert plain text into Fibo-Lite's transparent JSON-structured prompts - Bria's unique controllability layer that no closed model offers. Built for agentic and enterprise workflows.
Bria GenFill enables high-quality object addition or visual transformation. Trained exclusively on licensed data for safe and risk-free commercial use. Access the model's source code and weights: https://bria.ai/contact-us
The GenFill Route enables the generation of objects by prompt in a specific region of an image. You can define the area for object generation by using a mask that outlines the region where the object will be created. Our model is optimized to work seamlessly with blob-shaped masks.
Background removal you steer with a prompt. Starts from the same clean cut as Bria RMBG, then adds, drops or isolates objects from plain words like "the person with the mat and the plant" or "only the chair". Original pixels, real alpha, nothing repainted.
Upscale any image 2x or 4x, up to 8192×8192, with Bria Increase Resolution. Preserves the original content — no regeneration, no altered details. Commercial-safe
Bria Product Dimensions turns one product photo and its measurements into a marketplace-ready dimension image with callout lines, labels, and weight or capacity readouts
Place any product in any scenery with just a prompt or reference image while maintaining high integrity of the product. Trained exclusively on licensed data for safe and risk-free commercial use and optimized for eCommerce.
Structure Reference allows generating new images while preserving the structure of an input image, guided by text prompts. Perfect for transforming sketches, illustrations, or photos into new illustrations. Trained exclusively on licensed data for safe and risk-free commercial use.
Generate professional, eCommerce-ready product shots by replacing backgrounds with realistic lighting and accurate perspective from a simple text prompt. Trained exclusively on licensed data for safe commercial use.
RMBG-3.0 removes backgrounds and returns a transparent PNG. It holds up on hard cases: group shots, motion blur, mesh and openwork, shadows and reflections. It keeps product text and logos intact. Built on Fibo 2.0 and trained only on licensed data.
Bria's Text-to-Image model, trained exclusively on licensed data for safe and risk-free commercial use. Available also as source code and weights. For access to weights: https://bria.ai/contact-us
Bria's Text-to-Image model with perfect harmony of latency and quality. Trained exclusively on licensed data for safe and risk-free commercial use. Available also as source code and weights. For access to weights: https://bria.ai/contact-us
Bria's Text-to-Image model for HD images. Trained exclusively on licensed data for safe and risk-free commercial use. Available also as source code and weights. For access to weights: https://bria.ai/contact-us
Professional-grade creative upscaler that doubles resolution up to 10MP, regenerating sharper textures, refined details, and cleaner faces. Trained exclusively on licensed data for risk-free commercial use.
Automatically remove backgrounds from videos -perfect for creating clean, professional content without a green screen.
Remove background from videos filmed using chromakey, with automatic green spill suppression for clean, professional edges.
Remove video backgrounds in real time with Bria’s VRMBG 3.0 model. Built for live streaming, real-time video apps, content creation, and low-latency workflows that need fast, accurate background removal.
Remove backgrounds from any video with Bria's VRMBG 3.0. Fast, accurate background removal across talking heads, podcasts, product videos, commercials, and cinematic footage.
High-fidelity keypoint-driven video object removal - minimal input, strong temporal consistency. Trained on licensed data for risk-free commercial video editing.
High-fidelity mask-based video object removal with strong temporal consistency. Erase unwanted objects, people, or elements while preserving aesthetic quality. Trained on licensed data for risk-free commercial use.
Erase unwanted objects, people, or elements from video with a text prompt. High-fidelity output with strong temporal consistency, trained on licensed data for safe commercial use.
Professional-grade video upscaler with strong temporal consistency, enhancing videos up to 8K resolution. Trained on fully licensed and commercially safe data - risk-free for production and enterprise use.
Transfer motion from a video to characters in an image using Dreamactor v2. Great performance for non-human and multiple characters
OmniHuman generates video using an image of a human figure paired with an audio file. It produces vivid, high-quality videos where the character’s emotions and movements maintain a strong correlation with the audio.
Omnihuman v1.5 is a new and improved version of Omnihuman. It generates video using an image of a human figure paired with an audio file. It produces vivid, high-quality videos where the character’s emotions and movements maintain a strong correlation with the audio.
Omnihuman v2 generates video using an image of a human figure paired with an audio file. It produces vivid, high-quality videos where the character’s emotions and movements maintain a strong correlation with the audio.
Seed Audio 1.0 is a new audio model from Bytedance that can generate high-quality, natural sounding audio using text, reference audios or an image.
Seed Speech developed by ByteDance, is a family of large-scale text-to-speech models capable of synthesizing speech that is virtually indistinguishable from human speech.
Seed 2.0 Mini is a high-performance multimodal model optimized for low latency and high concurrency. It supports text, image, and video input with 256K context and configurable thinking/reasoning modes.
ByteDance's most advanced image-to-video model, fast tier. Lower latency and cost with synchronized audio, start and end frame control, and motion prompts.
ByteDance's most advanced reference-to-video model, fast tier. Lower latency and cost with up to 9 images, 3 videos, and 3 audio clips as inputs.
ByteDance's most advanced text-to-video model, fast tier. Lower latency and cost with cinematic output, native audio, multi-shot editing, and director-level camera control.
Seedance 2.0 Mini is a faster version of Seedance 2.0 that brings great performance and high generation speed at a lower cost.
Seedance 2.0 Mini is a faster version of Seedance 2.0 that brings great performance and high generation speed at a lower cost.
Seedance 2.0 Mini is a faster version of Seedance 2.0 that brings great performance and high generation speed at a lower cost.
US hosted version of ByteDance's most advanced image-to-video model. Animate still images into cinematic video with synchronized audio, start and end frame control, and motion prompts.
US hosted version of ByteDance's most advanced reference-to-video model. Generate video from up to 9 images, 3 videos, and 3 audio clips with native audio and cinematic camera control.
US hosted version of ByteDance's most advanced text-to-video model. Cinematic output with native audio, multi-shot editing, real-world physics, and director-level camera control.
Dreamina Seedance 2.5 Flash is the fast tier of Seedance 2.5. It animates a still image into a 4 to 30 second clip with native audio at 480p or 720p, with optional end-frame control.
Dreamina Seedance 2.5 Flash is the fast tier of Seedance 2.5. It generates, edits and extends video from up to 10 images, 5 videos and 5 audio references at 480p or 720p.
Dreamina Seedance 2.5 Flash is the fast tier of Seedance 2.5. It generates 4 to 30 second video with native audio from a text prompt at 480p or 720p.
Draft completion endpoint for Seedance 2.5 US - submit a draft id to regenerate the task at 1080p or 4K.
US-hosted ByteDance Seedance 2.5 animates still images with synchronized audio and optional end-frame control. Generate videos up to 30 seconds at 480p, 720p, 1080p or 4K.
US-hosted ByteDance Seedance 2.5 generates video with native audio from up to 30 images, 10 videos, and 10 audio references. Supports reference-guided generation, video editing, and extension at 480p, 720p, 1080p or 4K.
US-hosted ByteDance Seedance 2.5 generates cinematic video from text with synchronized audio, up to 30-second duration, and 480p, 720p, 1080p or 4K output.
Image to Video endpoint for Seedance 1.0 Pro Fast, a next-generation video model designed to deliver maximum performance at minimal cost
Text to Video endpoint for Seedance 1.0 Pro Fast, a next-generation video model designed to deliver maximum performance at minimal cost
Seedance 1.0 Pro, a high quality video generation model developed by Bytedance.
Seedance 1.0 Pro, a high quality video generation model developed by Bytedance.
A new-generation image creation model ByteDance, Seedream 4.5 integrates image generation and image editing capabilities into a single, unified architecture.
A new-generation image creation model ByteDance, Seedream 4.5 integrates image generation and image editing capabilities into a single, unified architecture.
A new-generation image creation model ByteDance, Seedream 4.0 integrates image generation and image editing capabilities into a single, unified architecture.
A new-generation image creation model ByteDance, Seedream 4.0 integrates image generation and image editing capabilities into a single, unified architecture.
Seedream 5.5 edits images from text instructions and up to 10 reference images, with configurable output dimensions and support for transparent PNG backgrounds.
Seedream 5.5 decomposes an image into a base image and independently editable PNG layers, with layer positions, stacking order, and optional text recognition.
Seedream 5.5 generates images from text prompts with configurable dimensions, batch generation, and support for transparent PNG backgrounds.
Seedream 5.0 Flash is a fast image generation and editing model, built for workflows where speed and budget matter.
Seedream 5.0 Flash is a fast image generation and editing model, built for workflows where speed and budget matter.
Seedream 5.0 Flash is a fast image generation and editing model, built for workflows where speed and budget matter.
Seedream 5.0 Pro is grounded, region-precise image editing model that changes one element while keeping the rest of the frame intact with layer separation, sketch completion, and up to 10 reference images.
Splits a finished image into independent, editable transparent-PNG layers — background plus separate elements, from a text description, returning 2 to 17 layers per call for non-destructive reuse in design tools.
ByteDance's Seedream 5.0 Pro is flagship text-to-image model, with deep-thinking prompt understanding, native text in 14 languages, and precise control over dense layouts and structured designs.
US inference endpoint for Seedream 5.0 Pro Image Editing.
US Inference endpoint for Seedream 5.0 Pro Text-to-Image.
Upscale videos with Bytedance's video upscaler.
CassetteAI’s model generates a 30-second sample in under 2 seconds and a full 3-minute track in under 10 seconds. At 44.1 kHz stereo audio, expect a level of professional consistency with no breaks, no squeaks, and no random interruptions in your creations.
Create stunningly realistic sound effects in seconds - CassetteAI's Sound Effects Model generates high-quality SFX up to 30 seconds long in just 1 second of processing time
Add sound effects to your videos
Whether you're working on memes, videos, games, or AI agents, Chatterbox brings your content to life. Use the first tts from resemble ai.
Whether you're working on memes, videos, games, or AI agents, Chatterbox brings your content to life. Use the first tts from resemble ai.
Whether you're working on memes, videos, games, or AI agents, Chatterbox brings your content to life. Use the first tts from resemble ai.
Clarity upscaler for upscaling images with high very fidelity.
An advanced image enhancement tool designed specifically for facial details and portrait photography, utilizing Clarity AI's upscaling technology.
Do high precision video upscaling that respects the original video perfectly using Crystal Upscaler's new video upscaling method!
Fix distorted or blurred photos of people with CodeFormer.
Cohere Transcribe turns your business audio into accurate text, ready for search, analytics, and automation
Generate high fidelity, studio quality videos of your avatar speaking or singing using the Aurora from Creatify team!
Create product, UGC, and presenter videos with synchronized native audio from text, with optional image and audio inputs.
Create creative upscaled images.
Bring colors into old or new black and white photos with DDColor.
Real-time, prompt-driven video editing over WebRTC. Restyle, swap backgrounds, and add or replace objects live on a webcam or streamed feed at interactive latency.
Real-time, prompt-driven video editing over WebRTC. Restyle, swap backgrounds, and add or replace objects live on a webcam or streamed feed at interactive latency.
Lucy-5B is a model that can create 5-second I2V videos in under 5 seconds, achieving >1x RTF end-to-end
Restyle videos up to 30 min long - maintaining maximum detail quality.
Realtime Try On experience with Decart Lucy 2.1 VTON
Enhance speech audio by removing background noise and upsampling to 48KHz
SOTA stemming model for voice, drums, bass, guitar and more.
Generates depth maps from video using Video Depth Anything (CVPR 2025). Produces per-frame depth estimation with temporal consistency across frames. Supports 3 model sizes (Small, Base, Large), 5 colormaps including grayscale, side-by-side comparison with the original video, and raw depth export as .npz. Useful for 3D reconstruction, video effects, compositing, and scene understanding.
Predict poses from images.
Predict poses from videos.
Isolate audio tracks using ElevenLabs advanced audio isolation technology.
Generate dubbed videos or audios using ElevenLabs Dubbing feature!
Align the transcript and your audio recording using Elevenlab's forced alignment feature!
Generate high quality, realistic music with fine controls using Elevenlabs Music!
Generate high quality, realistic music with fine controls using Elevenlabs Music v2!
Generate high quality, realistic music with fine controls using Elevenlabs Music v2.5!
Generate sound effects using ElevenLabs advanced sound effects model.
Generate sound effects up to 30 seconds from a text prompt with ElevenLabs SFX v3, with optional seamless looping.
Generate text from speech using ElevenLabs advanced speech-to-text model.
Use Scribe-V2 from ElevenLabs to do blazingly fast speech to text inferences!
Generate realistic audio dialogues using Eleven-v3 from ElevenLabs.
Generate text-to-speech audio using Eleven-v3 from ElevenLabs.
Generate expressive speech with Eleven v4 from ElevenLabs. Control delivery with audio tags, voice stability, similarity settings, and IPA pronunciation.
Generate speech with Eleven v4 Turbo from ElevenLabs. Choose a voice and control delivery with audio tags, stability, similarity settings, and IPA pronunciation.
Generate multilingual text-to-speech audio using ElevenLabs TTS Multilingual v2.
Generate high-speed text-to-speech audio using ElevenLabs TTS Turbo v2.5.
Change the voices in your audios with voices in ElevenLabs!
High-quality text-to-image model by Baidu. Supports English, Chinese, and Japanese prompts with built-in prompt expansion.
High-quality text-to-image model by Baidu. Supports English, Chinese, and Japanese prompts with built-in prompt expansion.
High-quality text-to-image model by Baidu. Supports English, Chinese, and Japanese prompts with built-in prompt expansion.
High-quality text-to-image model by Baidu. Supports English, Chinese, and Japanese prompts with built-in prompt expansion.
Upscale images by a given factor.
EVF-SAM2 combines natural language understanding with advanced segmentation capabilities, allowing you to precisely mask image regions using intuitive positive and negative text prompts.
F5 TTS
FASHN v1.5 delivers precise virtual try-on capabilities, accurately rendering garment details like text and patterns at 576x864 resolution from both on-model and flat-lay photo references.
FASHN v1.6 delivers precise virtual try-on capabilities, accurately rendering garment details like text and patterns at 864x1296 resolution from both on-model and flat-lay photo references.
Run SDXL at the speed of light
Run SDXL at the speed of light
Run SDXL at the speed of light
Run SDXL at the speed of light
Run SDXL at the speed of light
Run SDXL at the speed of light
FeyNobg is a state of the art AI model for background removal from feyninc
Compose videos from multiple media sources using FFmpeg API.
ffmpeg endpoint for first, middle and last frame extraction from videos
A fal.ai endpoint that stitches an ordered list of images into an MP4 video by holding each image for a specified number of frames at a configurable frame rate
Get EBU R128 loudness normalization from audio files using FFmpeg API.
Merge videos with standalone audio files or audio from video files.
Merge audios into a single audio using FFmpeg API!
Use ffmpeg capabilities to merge 2 or more videos.
Get encoding metadata from video and audio files using FFmpeg API.
Get waveform data from audio files using FFmpeg API.
Interpolate images with FILM - Frame Interpolation for Large Motion
Interpolate videos with FILM - Frame Interpolation for Large Motion
Finegrain Eraser removes objects—along with their shadows, reflections, and lighting artifacts—using only natural language, seamlessly filling the scene with contextually accurate content.
Finegrain Eraser removes any object selected with a bounding box—along with its shadows, reflections, and lighting artifacts—seamlessly reconstructing the scene with contextually accurate content.
Finegrain Eraser removes any object selected with a mask—along with its shadows, reflections, and lighting artifacts—seamlessly reconstructing the scene with contextually accurate content.
Florence-2 is an advanced vision foundation model that uses a prompt-based approach to handle a wide range of vision and vision-language tasks
Florence-2 is an advanced vision foundation model that uses a prompt-based approach to handle a wide range of vision and vision-language tasks
Florence-2 is an advanced vision foundation model that uses a prompt-based approach to handle a wide range of vision and vision-language tasks
Florence-2 is an advanced vision foundation model that uses a prompt-based approach to handle a wide range of vision and vision-language tasks
Florence-2 is an advanced vision foundation model that uses a prompt-based approach to handle a wide range of vision and vision-language tasks
Florence-2 is an advanced vision foundation model that uses a prompt-based approach to handle a wide range of vision and vision-language tasks
Florence-2 is an advanced vision foundation model that uses a prompt-based approach to handle a wide range of vision and vision-language tasks
Florence-2 is an advanced vision foundation model that uses a prompt-based approach to handle a wide range of vision and vision-language tasks
Florence-2 is an advanced vision foundation model that uses a prompt-based approach to handle a wide range of vision and vision-language tasks
Florence-2 is an advanced vision foundation model that uses a prompt-based approach to handle a wide range of vision and vision-language tasks
Florence-2 is an advanced vision foundation model that uses a prompt-based approach to handle a wide range of vision and vision-language tasks
Florence-2 is an advanced vision foundation model that uses a prompt-based approach to handle a wide range of vision and vision-language tasks
Florence-2 is an advanced vision foundation model that uses a prompt-based approach to handle a wide range of vision and vision-language tasks
Florence-2 is an advanced vision foundation model that uses a prompt-based approach to handle a wide range of vision and vision-language tasks
FLUX.1 [dev] is a 12 billion parameter flow transformer that generates high-quality images from text. It is suitable for personal and commercial use.
FLUX.1 [dev] is a 12 billion parameter flow transformer that generates high-quality images from text. It is suitable for personal and commercial use.
FLUX.1 [dev] Redux is a high-performance endpoint for the FLUX.1 [dev] model that enables rapid transformation of existing images, delivering high-quality style transfers and image modifications with the core FLUX capabilities.
FLUX.1 Krea [dev] is a 12 billion parameter flow transformer that generates high-quality images from text with incredible aesthetics. It is suitable for personal and commercial use.
FLUX.1 Krea [dev] is a 12 billion parameter flow transformer that generates high-quality images from text with incredible aesthetics. It is suitable for personal and commercial use.
FLUX.1 Krea [dev] Redux is a high-performance endpoint for the FLUX.1 Krea [dev] model that enables rapid transformation of existing images, delivering high-quality style transfers and image modifications with the core FLUX capabilities.
Fastest inference in the world for the 12 billion parameter FLUX.1 [schnell] text-to-image model.
FLUX.1 [schnell] Redux is a high-performance endpoint for the FLUX.1 [schnell] model that enables rapid transformation of existing images, delivering high-quality style transfers and image modifications with the core FLUX capabilities.
FLUX.1 SRPO [dev] is a 12 billion parameter flow transformer that generates high-quality images from text with incredible aesthetics. It is suitable for personal and commercial use.
FLUX.1 SRPO [dev] is a 12 billion parameter flow transformer that generates high-quality images from text with incredible aesthetics. It is suitable for personal and commercial use.
Text-to-image generation with FLUX.2 [dev] from Black Forest Labs. Enhanced realism, crisper text generation, and native editing capabilities.
Image-to-image editing with FLUX.2 [dev] from Black Forest Labs. Precise modifications using natural language descriptions and hex color control.
Text-to-image generation with FLUX.2 [dev] from Black Forest Labs. Enhanced realism, crisper text generation, and native editing capabilities— in a flash.
Image-to-image editing with FLUX.2 [dev] from Black Forest Labs. Precise modifications using natural language descriptions and hex color control—in a flash.
Text-to-image generation with FLUX.2 [flex] from Black Forest Labs. Features adjustable inference steps and guidance scale for fine-tuned control. Enhanced typography and text rendering capabilities.
Image editing with FLUX.2 [flex] from Black Forest Labs. Supports multi-reference editing with customizable inference steps and enhanced text rendering.
Text-to-image generation with FLUX.2 [klein] 4B from Black Forest Labs. Enhanced realism, crisper text generation, and native editing capabilities.
Text-to-image generation with FLUX.2 [klein] 4B Base from Black Forest Labs. Enhanced realism, crisper text generation, and native editing capabilities.
Image-to-image editing with FLUX.2 [klein] 4B Base from Black Forest Labs. Precise modifications using natural language descriptions and hex color control.
Image-to-image editing with LoRA support for FLUX.2 [klein] 4B Base from Black Forest Labs. Specialized style transfer and domain-specific modifications.
Text-to-image generation with LoRA support for FLUX.2 [klein] 4B Base from Black Forest Labs. Custom style adaptation and fine-tuned model variations.
Image-to-image editing with FLUX.2 [klein] 4B from Black Forest Labs. Precise modifications using natural language descriptions and hex color control.
Image-to-image editing with FLUX.2 [klein] 4B from Black Forest Labs and custom LoRA. Precise modifications using natural language descriptions and hex color control.
Text-to-image generation with FLUX.2 [klein] 4B from Black Forest Labs and custom LoRA. Enhanced realism, crisper text generation, and native editing capabilities.
Text-to-image generation with FLUX.2 [klein] 9B from Black Forest Labs. Enhanced realism, crisper text generation, and native editing capabilities.
Text-to-image generation with FLUX.2 [klein] 9B Base from Black Forest Labs. Enhanced realism, crisper text generation, and native editing capabilities.
Image-to-image editing with Flux 2 [klein] 9B Base from Black Forest Labs. Precise modifications using natural language descriptions and hex color control.
Image-to-image editing with LoRA support for FLUX.2 [klein] 9B Base from Black Forest Labs. Specialized style transfer and domain-specific modifications.
Text-to-image generation with LoRA support for FLUX.2 [klein] 9B Base from Black Forest Labs. Custom style adaptation and fine-tuned model variations.
Fine-tune FLUX.2 [klein] 9B from Black Forest Labs with custom datasets. Create specialized LoRA adaptations for specific editing tasks.
Fine-tune FLUX.2 [klein] 9B from Black Forest Labs with custom datasets. Create specialized LoRA adaptations for specific editing tasks.
Image-to-image editing with FLUX.2 [klein] 9B from Black Forest Labs. Precise modifications using natural language descriptions and hex color control.
Image-to-image editing with FLUX.2 [klein] 9B from Black Forest Labs and custom LoRA. Precise modifications using natural language descriptions and hex color control.
Text-to-image generation with FLUX.2 [klein] 9B from Black Forest Labs and custom LoRA.
Realtime generation with FLUX.2 [klein] from Black Forest Labs.
Text-to-image generation with LoRA support for FLUX.2 [dev] from Black Forest Labs. Custom style adaptation and fine-tuned model variations.
Image-to-image editing with LoRA support for FLUX.2 [dev] from Black Forest Labs. Specialized style transfer and domain-specific modifications.
Add a background to images with white/clean background
Virtually furnishes an empty apartment
Ballpoint pen sketch drawing style
Transforms images into comic book style
Extends a face into a full body portrait
HDR surrealistic effect with intense colors
Generates same object from different angles (azimuth/elevation)
Makes images more photorealistic and natural
Generates satellite/aerial view style images
Applies sepia vintage effect to images
Virtual clothing try-on (2 images: person + garment)
FLUX.2 [max] delivers state-of-the-art image generation and advanced image editing with exceptional realism, precision, and consistency.
FLUX.2 [max] delivers state-of-the-art image generation and advanced image editing with exceptional realism, precision, and consistency.
Image editing with FLUX.2 [pro] from Black Forest Labs. Ideal for high-quality image manipulation, style transfer, and sequential editing workflows
Text-to-image generation with FLUX.2 [pro] from Black Forest Labs. Optimized for maximum quality, exceptional photorealism and artistic images.
Outpainting generation with FLUX.2 [pro] from Black Forest Labs. Optimized for maximum quality, exceptional photorealism and artistic images.
Fine-tune FLUX.2 [dev] from Black Forest Labs with custom datasets. Create specialized LoRA adaptations for specific styles and domains.
Fine-tune FLUX.2 [dev] from Black Forest Labs with custom datasets. Create specialized LoRA adaptations for specific editing tasks.
Fine-tune FLUX.2 [dev] from Black Forest Labs with custom datasets. Create specialized LoRA adaptations for specific styles and domains.
Fine-tune FLUX.2 [dev] from Black Forest Labs with custom datasets. Create specialized LoRA adaptations for specific editing tasks.
Text-to-image generation with FLUX.2 [dev] from Black Forest Labs. Enhanced realism, crisper text generation, and native editing capabilities—all at turbo speed.
Image-to-image editing with FLUX.2 [dev] from Black Forest Labs. Precise modifications using natural language descriptions and hex color control—all at turbo speed.
FLUX.1 [dev] is a 12 billion parameter flow transformer that generates high-quality images from text. It is suitable for personal and commercial use.
FLUX.1 Image-to-Image is a high-performance endpoint for the FLUX.1 [dev] model that enables rapid transformation of existing images, delivering high-quality style transfers and image modifications with the core FLUX capabilities.
FLUX.1 [dev] Redux is a high-performance endpoint for the FLUX.1 [dev] model that enables rapid transformation of existing images, delivering high-quality style transfers and image modifications with the core FLUX capabilities.
A versatile endpoint for the FLUX.1 [dev] model that supports multiple AI extensions including LoRA, ControlNet conditioning, and IP-Adapter integration, enabling comprehensive control over image generation through various guidance methods.
A specialized FLUX endpoint combining differential diffusion control with LoRA, ControlNet, and IP-Adapter support, enabling precise, region-specific image transformations through customizable change maps.
FLUX General Image-to-Image is a versatile endpoint that transforms existing images with support for LoRA, ControlNet, and IP-Adapter extensions, enabling precise control over style transfer, modifications, and artistic variations through multiple guidance methods.
FLUX General Inpainting is a versatile endpoint that enables precise image editing and completion, supporting multiple AI extensions including LoRA, ControlNet, and IP-Adapter for enhanced control over inpainting results and sophisticated image modifications.
A general purpose endpoint for the FLUX.1 [dev] model, implementing the RF-Inversion pipeline. This can be used to edit a reference image based on a prompt.
Frontier image editing model.
Fast endpoint for the FLUX.1 Kontext [dev] model with LoRA support, enabling rapid and high-quality image editing using pre-trained LoRA adaptations for specific styles, brand identities, and product-specific outputs.
Fast inpainting endpoint for the FLUX.1 Kontext [dev] model with LoRA support, enabling rapid and high-quality image inpainting with reference images, while using pre-trained LoRA adaptations for specific styles, brand identities, and product-specific outputs.
Super fast text-to-image endpoint for the FLUX.1 Kontext [dev] model with LoRA support, enabling rapid and high-quality image generation using pre-trained LoRA adaptations for personalization, specific styles, brand identities, and product-specific outputs.
FLUX.1 Krea [dev] is a 12 billion parameter flow transformer that generates high-quality images from text with incredible aesthetics. It is suitable for personal and commercial use.
FLUX.1 Krea [dev] is a 12 billion parameter flow transformer that generates high-quality images from text with incredible aesthetics. It is suitable for personal and commercial use.
Super fast endpoint for the FLUX.1 [dev] model with LoRA support, enabling rapid and high-quality image generation using pre-trained LoRA adaptations for personalization, specific styles, brand identities, and product-specific outputs.
FLUX LoRA Image-to-Image is a high-performance endpoint that transforms existing images using FLUX models, leveraging LoRA adaptations to enable rapid and precise image style transfer, modifications, and artistic variations.
Super fast endpoint for the FLUX.1 [dev] inpainting model with LoRA support, enabling rapid and high-quality image inpaingting using pre-trained LoRA adaptations for personalization, specific styles, brand identities, and product-specific outputs.
Super fast endpoint for the FLUX.1 [dev] model with LoRA support, enabling rapid and high-quality image generation using pre-trained LoRA adaptations for personalization, specific styles, brand identities, and product-specific outputs.
FLUX.1 Krea [dev] Redux is a high-performance endpoint for the FLUX.1 Krea [dev] model that enables rapid transformation of existing images, delivering high-quality style transfers and image modifications with the core FLUX capabilities.
Super fast endpoint for the FLUX.1 [dev] model with LoRA support, enabling rapid and high-quality image generation using pre-trained LoRA adaptations for personalization, specific styles, brand identities, and product-specific outputs.
Generate high-quality images from depth maps using Flux.1 [dev] depth estimation model. The model produces accurate depth representations for scene understanding and 3D visualization.
Train styles, people and other subjects at blazing speeds.
FLUX.1 [dev] Fill is a high-performance endpoint for the FLUX.1 [pro] model that enables rapid transformation of existing images, delivering high-quality style transfers and image modifications with the core FLUX capabilities.
FLUX LoRA Image-to-Image is a high-performance endpoint that transforms existing images using FLUX models, leveraging LoRA adaptations to enable rapid and precise image style transfer, modifications, and artistic variations.
Super fast endpoint for the FLUX.1 [dev] inpainting model with LoRA support, enabling rapid and high-quality image inpaingting using pre-trained LoRA adaptations for personalization, specific styles, brand identities, and product-specific outputs.
FLUX LoRA training optimized for portrait generation, with bright highlights, excellent prompt following and highly detailed results.
Super fast endpoint for the FLUX.1 [dev] model with LoRA support, enabling rapid and high-quality image generation using pre-trained LoRA adaptations for personalization, specific styles, brand identities, and product-specific outputs.
FLUX.1 Kontext [pro] handles both text and reference images as inputs, seamlessly enabling targeted, local edits and complex transformations of entire scenes.
FLUX.1 Kontext [max] is a model with greatly improved prompt adherence and typography generation meet premium consistency for editing without compromise on speed.
Experimental version of FLUX.1 Kontext [max] with multi image handling capabilities
FLUX.1 Kontext [max] text-to-image is a new premium model brings maximum performance across all aspects – greatly improved prompt adherence.
Experimental version of FLUX.1 Kontext [pro] with multi image handling capabilities
The FLUX.1 Kontext [pro] text-to-image delivers state-of-the-art image generation results with unprecedented prompt following, photorealistic rendering, and flawless typography.
FLUX1.1 [pro] is an enhanced version of FLUX.1 [pro], improved image generation capabilities, delivering superior composition, detail, and artistic fidelity compared to its predecessor.
FLUX1.1 [pro] Redux is a high-performance endpoint for the FLUX1.1 [pro] model that enables rapid transformation of existing images, delivering high-quality style transfers and image modifications with the core FLUX capabilities.
FLUX1.1 [pro] ultra is the newest version of FLUX1.1 [pro], maintaining professional-grade image quality while delivering up to 2K resolution with improved photo realism.
FLUX1.1 [pro] ultra fine-tuned is the newest version of FLUX1.1 [pro] with a fine-tuned LoRA, maintaining professional-grade image quality while delivering up to 2K resolution with improved photo realism.
FLUX1.1 [pro] ultra Redux is a high-performance endpoint for the FLUX1.1 [pro] model that enables rapid transformation of existing images, delivering high-quality style transfers and image modifications with the core FLUX capabilities.
Latest object erasing model from Black Forest Labs. Remove undesired objects, texts from images.
FLUX.1 [pro] Fill is a high-performance endpoint for the FLUX.1 [pro] model that enables rapid transformation of existing images, delivering high-quality style transfers and image modifications with the core FLUX capabilities.
FLUX.1 [pro] Fill Fine-tuned is a high-performance endpoint for the FLUX.1 [pro] model with a fine-tuned LoRA that enables rapid transformation of existing images, delivering high-quality style transfers and image modifications with the core FLUX capabilities.
Generate virtual try-on results from a person image plus one or more garment references.
An endpoint for personalized image generation using Flux as per given description.
FLUX.1 [schnell] is a 12 billion parameter flow transformer that generates high-quality images from text in 1 to 4 steps, suitable for personal and commercial use.
FLUX.1 [schnell] Redux is a high-performance endpoint for the FLUX.1 [schnell] model that enables rapid transformation of existing images, delivering high-quality style transfers and image modifications with the core FLUX capabilities.
FLUX.1 SRPO [dev] is a 12 billion parameter flow transformer that generates high-quality images from text with incredible aesthetics. It is suitable for personal and commercial use.
FLUX.1 SRPO [dev] is a 12 billion parameter flow transformer that generates high-quality images from text with incredible aesthetics. It is suitable for personal and commercial use.
Framepack is an efficient Image-to-video model that autoregressively generates videos.
Framepack is an efficient Image-to-video model that autoregressively generates videos.
Framepack is an efficient Image-to-video model that autoregressively generates videos.
Google's famous original image generation and editing model, a.k.a Nano Banana
Google's famous original image generation and editing model, a.k.a Nano Banana
Gemini 3.1 Flash Image (a.k.a Nano Banana 2) is Google's new state-of-the-art fast image generation and editing model
Gemini 3.1 Flash Image (a.k.a. Nano Banana 2) is Google's new state-of-the-art fast image generation and editing model
Newest audio model from Google introduces granular audio tags that give you precise control to direct AI speech for expressive audio generation.
Gemini 3 Pro Image (a.k.a Nano Banana Pro) is Google's state-of-the-art high-fidelity image generation and editing model
Gemini 3 Pro Image (a.k.a Nano Banana Pro) is Google's state-of-the-art high-fidelity image generation and editing model
Use Gemini TTS Models to convert your prompts to real audio.
Create high-quality images with accurate text rendering and rich knowledge details—supports editing, style transfer, and maintaining consistent characters across multiple images.
Create high-quality images with accurate text rendering and rich knowledge details—supports editing, style transfer, and maintaining consistent characters across multiple images.
Generate expressive speech with Gemini 3.8 Flash Lite TTS. Choose from 30 voices, guide delivery with style instructions, and create single-speaker narration or two-speaker dialogue.
Generate expressive speech with Gemini 3.8 Flash TTS. Choose from 30 voices, direct delivery with style instructions, and create single-speaker narration or two-speaker dialogue.
Creates video with synchronized audio from text input. Grounded in Gemini's real-world knowledge, with improved physics understanding for more coherent motion and interaction.
Edits generated video across multiple conversational turns while preserving scene coherence. Applies iterative changes through natural-language instructions without regenerating the full sequence from scratch.
Animates a still image into video with audio. Extends a single frame into coherent motion, grounded in Gemini's physical understanding of how scenes and subjects behave.
Generates video with audio from combined multimodal references. Accepts text, images, audio, and video together as input to guide subject, motion, style, and sound in the output.
Gemini Omni Flash 1.1 is Google's multimodal video model. This endpoint edits video through natural-language instruction, applying the requested change while preserving the parts of the scene you want kept, and carrying character and scene consistency across successive edits.
Gemini Omni Flash 1.1 is Google's multimodal video model. This endpoint animates a still image into video with synchronized audio, extending a single frame into coherent motion that reflects the logic of the real world.
Gemini Omni Flash 1.1 is Google's multimodal video model. This endpoint generates video from combined multimodal references, images, videos and text together. Reasoning across all inputs to produce a single coherent result, with characters retaining their face, clothing, and voice throughout
Gemini Omni Flash 1.1 is Google's multimodal video model. This endpoint generates video with synchronized native audio from a text prompt, grounded in Gemini's real-world knowledge and physics understanding, with cinematic camera control expressed in natural language.
Gemini Omni Flash 1.2 is Google's multimodal video model. This endpoint edits an existing video of up to 15 seconds from a natural-language instruction, applying the requested change while keeping the rest of the scene unchanged. Output resolution is selectable from 360p, 720p, 1080p, or 4K.
Gemini Omni Flash 1.2 is Google's multimodal video model. This endpoint continues an existing video from a text prompt, conditioning on its last 10 seconds and appending a new 3 to 10 second segment. Output resolution is selectable from 360p, 720p, 1080p, or 4K.
Gemini Omni Flash 1.2 is Google's multimodal video model. This endpoint animates a start image into video with synchronized native audio, with an optional end image for first-to-last-frame interpolation of up to 10 seconds. Output is available at 360p, 720p, 1080p, or 4K, with durations from 3 to 15 seconds.
Gemini Omni Flash 1.2 is Google's multimodal video model. This endpoint generates video from a text prompt combined with up to ten reference images and up to three reference videos, each addressable in the prompt, using them as visual references for subjects and settings. Output is available at 360p, 720p, 1080p, or 4K, with durations from 3 to 15 seconds.
Gemini Omni Flash 1.2 is Google's multimodal video model. This endpoint generates video with synchronized native audio from a text prompt, at 360p, 720p, 1080p, or 4K, in 16:9 or 9:16, with durations from 3 to 15 seconds.
Gemini Omni Flash 1.2 is Google's multimodal video model. This endpoint upscales a video of up to 15 seconds to a selected resolution of up to 4K, with an optional text instruction to guide the result.
Lyria 3.5 is Google DeepMind's latest music generation model, and you can generate almost any type of music with it
Nano Banana 2.1 by Google generates images from text prompts, with output resolutions up to 4K, adjustable aspect ratios, and optional web search grounding.
Nano Banana 2.1 Edit by Google transforms reference images using text instructions, with output resolutions up to 4K and optional video or PDF context.
Nano banana lite is the efficiency-focused model in the image generation family. Sub-2 second latency with cost-effective generation and editing, fast multi-turn local edits, and 14 supported aspect ratios.
Nano banana lite is the efficiency-focused model in the image generation family. Sub-2 second latency with cost-effective generation and editing, fast multi-turn local edits, and 14 supported aspect ratios.
Generate realistic virtual try-on images from a person image and a clothing product image.
GPT Image 1.5 generates high-fidelity images with strong prompt adherence, preserving composition, lighting, and fine-grained detail.
GPT Image 1.5 generates high-fidelity images with strong prompt adherence, preserving composition, lighting, and fine-grained detail.
GPT Image 1 mini combines OpenAI's advanced language capabilities, powered by GPT-5, with GPT Image 1 Mini for efficient image generation.
GPT Image 1 mini combines OpenAI's advanced language capabilities, powered by GPT-5, with GPT Image 1 Mini for efficient image generation.
HeyGen Avatar 6 generates a talking video of a stock HeyGen avatar speaking a text script in a selected HeyGen voice, or lip-synced to an uploaded audio file. Output is 1080p or 2K in 16:9, 9:16, 1:1, 4:5 or 5:4.
HeyGen Avatar 6 animates a photo of a person into a talking video lip-synced to a supplied speech track, with the audio setting the video length. Output is 1080p or 2K, and an optional motion prompt directs expression and movement.
HeyGen Avatar 6 animates an animated or illustrated character from a reference video into a talking video lip-synced to a supplied speech track, with an optional reference photo of the same character. Videos of real people are not supported. Output is 1080p or 2K.
HeyGen Avatar 6 Turbo, the lower-cost tier, generates a talking video of a stock HeyGen avatar speaking a text script in a selected HeyGen voice, or lip-synced to an uploaded audio file. Output is 1080p or 2K in 16:9, 9:16, 1:1, 4:5 or 5:4.
HeyGen Avatar 6 Turbo, the lower-cost tier, animates a photo of a person into a talking video lip-synced to a supplied speech track, with the audio setting the video length. Output is 1080p or 2K, and an optional motion prompt directs expression and movement.
HeyGen Avatar 6 Turbo, the lower-cost tier, animates an animated or illustrated character from a reference video into a talking video lip-synced to a supplied speech track, with an optional reference photo of the same character. Videos of real people are not supported. Output is 1080p or 2K.
Heygen Avatar V3 Model for Digital Twin
Heygen Avatar 4 Digital Twin Model
Heygen Photo Avatar 4 Model
Create natural HeyGen Avatar V digital twin videos from text or audio, with lip-sync, optional backgrounds, captions, and MP4/WebM output.
HeyGen Video animates your image as the first frame of a 5 to 15 second clip, with generated sound, built on MiniMax H3 and post-trained by HeyGen.
HeyGen Video builds a new 5 to 15 second scene with sound that keeps the products, people or places from your reference images, videos or audio.
HeyGen Video, built on MiniMax H3 and post-trained by HeyGen, turns a prompt into a 5 to 15 second clip with its own sound.
HeyGen Voice turns text into natural speech with a library voice or an instant clone from a reference clip.
HeyGen Voice Turbo is the fast, low-cost tier of HeyGen Voice, with library voices and instant cloning from a reference clip.
Heygen Translate Model with Extreme Precision
Heygen Translate Model with Extreme Speed
Heygen Text to Video Generation Model
Use Heygen's Latest Model for Filler Word Removal.
Replace or dub audio on an existing video with high-accuracy avatar-inference lip-sync.
Replace or dub audio on an existing video with fast audio-only lip-sync.
Generate videos with a single prompt. Describe what you want in plain text, and the agent handles avatar selection, scripting, scene composition - all in one.
HiDream-I1 fast is a new open-source image generative foundation model with 17B parameters that achieves state-of-the-art image generation quality within 16 steps.
HiDream-I1 full is a new open-source image generative foundation model with 17B parameters that achieves state-of-the-art image generation quality within seconds.
HiDream-I1 full is a new open-source image generative foundation model with 17B parameters that achieves state-of-the-art image generation quality within seconds.
Unified image generation with HiDream-O1-Image. Create, edit, and personalize high-resolution images up to 2K—single native model handles text-to-image, editing, and custom subjects without external components.
Unified image generation with HiDream-O1-Image. Create, edit, and personalize high-resolution images up to 2K—single native model handles text-to-image, editing, and custom subjects without external components.
Unified image generation with HiDream-O1-Image. Create, edit, and personalize high-resolution images up to 2K—single native model handles text-to-image, editing, and custom subjects without external components.
Unified image generation with HiDream-O1-Image. Create, edit, and personalize high-resolution images up to 2K—single native model handles text-to-image, editing, and custom subjects without external components.
Generate 3D models from a single image with Hi3D.
Generate a 3D relief depth map with Hi3D from a single image.
Generate 3D models from multiple view images using Hi3D.
Convert a textured 3D model into a multicolor model suited for multicolor 3D printing with Hi3D.
Split a 3D model into parts with Hi3D.
Texture an existing geometry mesh using a reference image with Hi3D.
Generate 3D models from a single image with Hi3D V3.0.
Generate 3D models from multiple view images using Hi3D V3.0.
Generate 3D models from images with Hunyuan 3D Pro
Generate 3D models from text prompts with Hunyuan 3D Pro
Rapidly generate 3D models from images using Hunyuan 3D.
Create detailed, fully-textured 3D models with text
Use the amazing capabilities of hunyuan image 2.1 to generate images that express the feelings of your text.
Image editing endpoint for Hunyuan Image 3.0 Instruct.
Instruct version of Hunyuan-Image 3.0, with internal reasoning capabilities.
Leverage the state-of-the-art capabilities of Hunyuan Image 3.0 to generate visual content that effectively conveys the messaging of your written material.
Generate 3D human motions via text-to-generation interface of Hunyuan Motion!
Generate 3D human motions via text-to-generation interface of Hunyuan Motion!
Generate 3D models from your images using Hunyuan 3D. A native 3D generative model enabling versatile and high-quality 3D asset creation.
Generate 3D models from your images using Hunyuan 3D. A native 3D generative model enabling versatile and high-quality 3D asset creation.
Generate 3D models from your images using Hunyuan 3D. A native 3D generative model enabling versatile and high-quality 3D asset creation.
Generate 3D models from your images using Hunyuan 3D. A native 3D generative model enabling versatile and high-quality 3D asset creation.
Generate 3D models from your images using Hunyuan 3D. A native 3D generative model enabling versatile and high-quality 3D asset creation.
Generate 3D models from your images using Hunyuan 3D. A native 3D generative model enabling versatile and high-quality 3D asset creation.
Transform your photos into ultra-high-resolution 3D models in seconds. Film-quality geometry with PBR textures, ready for games, e-commerce, and 3D printing.
Turn simple sketches into detailed, fully-textured 3D models. Instantly convert your concept designs into formats ready for Unity, Unreal, and Blender.
Hyper3D Low Poly generates low-poly 3D models from reference images, with optional texture generation and support for triangle or quad topology.
Hyper3D Low Poly converts existing high-poly models into low-poly meshes with triangle or quad topology, optionally projecting and baking the original textures onto the resulting mesh.
Hyper3D Low Poly generates low-poly 3D models from text prompts, with optional texture generation and support for triangle or quad topology.
Hyper3D Rigging automatically rigs existing 3D models with a skeleton for animation.
Lists available animation presets for Hyper3D Rigging, returning animation IDs and metadata for use in rigging requests.
Rodin by Hyper3D generates realistic and production ready 3D models from text or images.
Rodin by Hyper3D generates realistic and production ready 3D models from text or images.
Rodin V2.5 by Hyper3D generates realistic and production ready 3D models from text or images.
Rodin V2.5 by Hyper3D generates realistic and production ready 3D models from text or images. Do fast prototyping using the fast model.
Rodin V2.5 by Hyper3D generates realistic and production ready 3D models from text or images.
Rodin V2.5 by Hyper3D generates realistic and production ready 3D models from text or images. Do fast prototyping using the fast model.
An endpoint for re-lighting photos and changing their backgrounds per a given description
Generate consistent character appearances across multiple images. Maintain facial features, proportions, and distinctive traits for cohesive storytelling and branding
Modify consistent characters while preserving their core identity. Edit poses, expressions, or clothing without losing recognizable character features
Transform your consistent character into different art styles, settings, or scenarios while maintaining their distinctive appearance and identity
Prompt-free object removal from an image and mask, erasing objects with their shadows and reflections and reconstructing the scene cleanly.
Remove backgrounds from existing images with Ideogram's remove background feature. Isolate subjects cleanly for compositing and creative reuse.
Ideogram Upscale enhances the resolution of the reference image by up to 2X and might enhance the reference image too. Optionally refine outputs with a prompt for guided improvements.
Generate high-quality images, posters, and logos with Ideogram V2. Features exceptional typography handling and realistic outputs optimized for commercial and creative use.
Transform existing images with Ideogram V2's editing capabilities. Modify, adjust, and refine images while maintaining high fidelity and realistic outputs with precise prompt control.
Reimagine existing images with Ideogram V2's remix feature. Create variations and adaptations while preserving core elements and adding new creative directions through prompt guidance.
Accelerated image generation with Ideogram V2 Turbo. Create high-quality visuals, posters, and logos with enhanced speed while maintaining Ideogram's signature quality.
Edit images faster with Ideogram V2 Turbo. Quick modifications and adjustments while preserving the high-quality standards and realistic outputs of Ideogram.
Rapidly create image variations with Ideogram V2 Turbo Remix. Fast and efficient reimagining of existing images while maintaining creative control through prompt guidance.
Generate high-quality images, posters, and logos with Ideogram V2A. Features exceptional typography handling and realistic outputs optimized for commercial and creative use.
Create variations of existing images with Ideogram V2A Remix while maintaining creative control through prompt guidance.
Accelerated image generation with Ideogram V2A Turbo. Create high-quality visuals, posters, and logos with enhanced speed while maintaining Ideogram's signature quality.
Rapidly create image variations with Ideogram V2A Turbo Remix. Fast and efficient reimagining of existing images while maintaining creative control through prompt guidance.
Generate high-quality images, posters, and logos with Ideogram V3. Features exceptional typography handling and realistic outputs optimized for commercial and creative use.
Transform existing images with Ideogram V3's editing capabilities. Modify, adjust, and refine images while maintaining high fidelity and realistic outputs with precise prompt control.
Generate images with transparent backgrounds using Ideogram Transparent model
Ideogram Layerize takes an existing flat graphic, removes text, and returns structured text containers you can edit/recompose in html or json format.
Extend existing images with Ideogram V3's reframe feature. Create expanded versions and adaptations while preserving main image and adding new creative directions through prompt guidance.
Reimagine existing images with Ideogram V3's remix feature. Create variations and adaptations while preserving core elements and adding new creative directions through prompt guidance.
Replace backgrounds existing images with Ideogram V3's replace background feature. Create variations and adaptations while preserving core elements and adding new creative directions through prompt guidance.
Generate high-quality images, posters, and logos with Ideogram's latest V4.0q — producing crisp visuals with accurate text rendering, fine detail, and full creative control for polished, ready-to-use designs.
Generate high-quality images, posters, and logos with Ideogram 4.5, with accurate text rendering and low, medium, or high quality tiers.
Edit images with Ideogram 4.5: prompt edits with up to 4 reference images, optional mask, and a high-precision mode that keeps unchanged pixels intact.
Ideogram 4.6 generates images from text prompts with configurable quality, image dimensions, and optional transparent backgrounds.
Ideogram 4.6 edits an image from a text instruction, with optional masks, reference images, and precision controls.
Generate high-quality images, posters, and logos with Ideogram's latest V4.0q — producing crisp visuals with accurate text rendering, fine detail, and full creative control for polished, ready-to-use designs IN A SECOND.
Ideogram V4.0q Image-to-Image transforms an input image with a text prompt, restyling and reworking the composition while preserving its core structure for prompt-faithful, high-fidelity edits.
Ideogram V4.0q Image-to-Image LoRA applies a custom-trained LoRA on top of an input image, steering edits toward a specific style, subject, or brand identity while keeping the source composition intact.
Generate high-quality images, posters, and logos with Ideogram's latest V4.0q — producing crisp visuals with accurate text rendering, fine detail, and full creative control for polished, ready-to-use designs FRACTION OF A SECOND.
Generate high-quality images, posters, and logos with Ideogram's latest V4.0q using LoRA — producing crisp visuals with accurate text rendering, fine detail, and full creative control for polished, ready-to-use designs.
Ideogram V4.0q Tiling generates seamless, edge-matching textures and patterns that repeat infinitely in any direction, ideal for backgrounds, surfaces, and wallpapers.
Ideogram V4.0q Tiling LoRA produces seamless repeatable patterns guided by a custom-trained LoRA, locking a specific aesthetic or motif into tileable textures for cohesive, large-scale surface design.
Train custom LoRAs for personalization, styles or other use cases on top of Ideogram V4.
Converts images into editable SVG vector graphics, with a default mode for finer details and a simple mode for fewer shapes.
Generates editable SVG vector graphics from text prompts, with optional reference images, three quality levels, and aspect ratio controls.
Create illusions conditioned on image.
Modify a face to look younger or older while keeping identity realistic.
Place a person’s photo into iconic cities worldwide.
Change facial expressions in photos with realistic results.
Change hairstyles and hair colors in photos realistically.
Generate professional headshot photos with customizable backgrounds.
Apply realistic makeup styles with adjustable intensity.
Remove unwanted objects seamlessly from any image.
Directional outpainting. Choose edges to expand. left, right, top, or center (uniform all sides). Only expanded areas are generated; an optional zoom-out pulls the frame back by the chosen amount.
Easily adjust the perspective of any image to different angles.
Restore old or damaged photos by fixing colors, scratches, and resolution.
Apply diverse photography styles and effects to transform your images.
Enhance and refine portrait photos with improved clarity and detail.
Place products naturally in a person’s hands for realistic marketing visuals.
Generate professional product photography with realistic lighting and backgrounds.
Adjust and enhance images with different lighting styles.
Apply artistic styles like impressionism, cubism, or surrealism to your images.
Transform objects with different surface textures like marble, wood, or fabric.
Try on clothes virtually by combining person and clothing images.
See how you or others might look at different ages, from younger to older, while preserving core facial features.
Transform any person into their baby version, while preserving the original pose and expression with childlike features.
Replace your photo's background with any scene you desire, from beach sunsets to urban landscapes, with perfect lighting and shadows
Transform your character's hair into broccoli style while keeping the original characters likeness
Transform your photos into vibrant cool cartoons with bold outlines and rich colors.
Perfect your photos with professional color grading, balanced tones, and vibrant yet natural colors
Change facial expressions in photos to any emotion you desire, from smiles to serious looks.
Enhance facial features with professional retouching while maintaining a natural, realistic look
Experiment with different hairstyles, from bald to any style you can imagine, while maintaining natural lighting and realistic results.
Remove unwanted objects or people from your photos while seamlessly blending the background.
Restore and enhance old or damaged photos by removing imperfections, adding color while preserving the original character and details of the image.
Transform your photos into cool plushies while keeping the original characters likeness
Turn your casual photos into stunning professional studio portraits with perfect lighting and high-end photography style.
Add details to faces, enhance face features, remove blur.
The reframe endpoint intelligently adjusts an image's aspect ratio while preserving the main subject's position, composition, pose, and perspective
Retouch photos of faces. Remove blemishes and improve the skin.
Place your subject in any scene you imagine, from enchanted forests to urban settings, with professional composition and lighting
Transform your photos into artistic masterpieces inspired by famous styles like Van Gogh's Starry Night or any artistic style you choose.
Remove all text and writing from images while preserving the background and natural appearance.
Transform your photos to any time of day, from golden hour to midnight, with appropriate lighting and atmosphere.
Add realistic weather effects like snowfall, rain, or fog to your photos while maintaining the scene's mood.
Transform your photos into wojak style while keeping the original characters likeness
Generate YouTube thumbnails with custom text
Depth Anything v2 preprocessor.
Holistically-Nested Edge Detection (HED) preprocessor.
Line art preprocessor.
MiDaS depth estimation preprocessor.
M-LSD line segment detection preprocessor.
PIDI (Pidinet) preprocessor.
Segment Anything Model (SAM) preprocessor.
Scribble preprocessor.
TEED (Temporal Edge Enhancement Detection) preprocessor.
ZoeDepth preprocessor.
Image2SVG transforms raster images into clean vector graphics, preserving visual quality while enabling scalable, customizable SVG outputs with precise control over detail levels.
Create depth maps using Midas depth estimation.
Create depth maps using Marigold depth estimation.
Predict the probability of an image being NSFW.
Remove the background from an image.
Generate natural, clear speeches using Index TTS 2.0 from IndexTeam
Text to Speech Endpoint for Inworld's TTS-1.5 Max.
Kandinsky 6.0 Pro turns a start frame into a high-resolution, cinematic video clip with controllable motion and strong visual fidelity.
Kandinsky 6.0 Pro is Kandinsky Lab's flagship text-to-video model, generating high-resolution clips with cinematic motion and precise prompt adherence.
Perform precise image edits using strong reference control, transforming subjects, styles, and local details while preserving visual consistency.
Kling Omni 3: Top-tier image-to-image with flawless consistency.
Kling Omni 3: Top-tier text-to-image with flawless consistency.
Kling Image V3: Latest kling image model
Kling V3: Latest Kling Image model
Kling 4.0 Flash animates a starting image into a 720p video, with optional text prompts and durations from 3 to 20 seconds.
Kling 4.0 Flash generates 720p videos from text prompts and image or video references, with durations from 3 to 20 seconds.
Kling 4.0 Flash generates videos from text prompts at 720p, with durations from 3 to 20 seconds and configurable aspect ratios.
Kling 4.0 Flash edits an input video using text instructions and optional image, video or element references, producing 3 to 20-second videos at 720p.
Kling Image V4 image-to-image endpoint.
Kling 4.0 animates a starting image into a 3 to 30-second video at 720p, 1080p or native 4K, with optional 10-bit HDR at 1080p and 4K, optional ending frames and per-shot prompts.
Generates long-form visual designs for webpages and ecommerce from a text prompt, with optional reference images and controls for length and resolution.
Kling 4.0 generates videos from text prompts and image or video references at 720p, 1080p or native 4K, with optional 10-bit HDR at 1080p and 4K, durations from 3 to 30 seconds and per-shot prompts.
Kling Image V4 text-to-image endpoint.
Kling 4.0 generates videos from text prompts at 720p, 1080p or native 4K, with optional 10-bit HDR at 1080p and 4K, durations from 3 to 30 seconds and support for per-shot prompts.
Kling 4.0 edits an input video using text instructions and optional image references, producing 3 to 30-second videos at 720p, 1080p or native 4K, with optional 10-bit HDR at 1080p and 4K.
Kling AI Avatar v2 Pro: The premium endpoint for creating avatar videos with realistic humans, animals, cartoons, or stylized characters
Kling AI Avatar v2 Standard: Endpoint for creating avatar videos with realistic humans, animals, cartoons, or stylized characters
Create Voices to be used with Kling Models Voice Control
Kling LipSync is an audio-to-video model that generates realistic lip movements from audio input.
Kling LipSync is a text-to-video model that generates realistic lip movements from text input.
Generate a video by taking a start frame and an end frame, animating the transition between them while following text-driven style and scene guidance.
Transform images, elements, and text into consistent, high-quality video scenes, ensuring stable character identity, object details, and environments.
Generate a video by taking a start frame and an end frame, animating the transition between them while following text-driven style and scene guidance.
Transform images, elements, and text into consistent, high-quality video scenes, ensuring stable character identity, object details, and environments.
Edit an existing video using natural-language instructions, transforming subjects, settings, and style while retaining the original motion structure.
Kling O1 Omni generates new shots guided by an input reference video, preserving cinematic language such as motion, and camera style to produce seamless scene continuity.
Edit an existing video using natural-language instructions, transforming subjects, settings, and style while retaining the original motion structure.
Kling O1 Omni generates new shots guided by an input reference video, preserving cinematic language such as motion, and camera style to produce seamless scene continuity.
Kling's Native 4K is a video generation model that directly outputs professional-grade 4K video in one step, eliminating the need for post-production upscaling
Kling's Native 4K is a video generation model that directly outputs professional-grade 4K video in one step, eliminating the need for post-production upscaling
Kling's Native 4K is a video generation model that directly outputs professional-grade 4K video in one step, eliminating the need for post-production upscaling
Kling's Native 4K is a video generation model that directly outputs professional-grade 4K video in one step, eliminating the need for post-production upscaling
Kling's Native 4K is a video generation model that directly outputs professional-grade 4K video in one step, eliminating the need for post-production upscaling
Generate a video by taking a start frame and an end frame, animating the transition between them while following text-driven style and scene guidance.
Transform images, elements, and text into consistent, high-quality video scenes, ensuring stable character identity, object details, and environments.
Generate realistic videos using Kling O3 from Kling Team!
Edit videos using Kling O3 from Kling Team!
Kling O3 Omni generates new shots guided by an input reference video, preserving cinematic language such as motion, and camera style to produce seamless scene continuity.
Generate a video by taking a start frame and an end frame, animating the transition between them while following text-driven style and scene guidance.
Transform images, elements, and text into consistent, high-quality video scenes, ensuring stable character identity, object details, and environments.
Generate realistic videos using Kling O3 from Kling Team!
Edit videos using Kling O3 from Kling Team!
Kling O3 Omni generates new shots guided by an input reference video, preserving cinematic language such as motion, and camera style to produce seamless scene continuity.
Generate video clips from your prompts using Kling 1.5 (pro)
Generate video clips from your prompts using Kling 1.6 (pro)
Generate video clips from your prompts using Kling 1.6 (std)
Kling AI Avatar Pro: The premium endpoint for creating avatar videos with realistic humans, animals, cartoons, or stylized characters
Kling AI Avatar Standard: Endpoint for creating avatar videos with realistic humans, animals, cartoons, or stylized characters
Generate video clips from your prompts using Kling 1.0
Generate speech from text prompts and different voices using the Kling TTS model, which leverages advanced AI techniques to create high-quality text-to-speech.
Kling 2.5 Turbo Pro: Top-tier image-to-video generation with unparalleled motion fluidity, cinematic visuals, and exceptional prompt precision.
Kling 2.5 Turbo Pro: Top-tier text-to-video generation with unparalleled motion fluidity, cinematic visuals, and exceptional prompt precision.
Kling 2.5 Turbo Standard: Top-tier image-to-video generation with unparalleled motion fluidity, cinematic visuals, and exceptional prompt precision.
Kling 2.6 Pro: Top-tier image-to-video with cinematic visuals, fluid motion, and native audio generation.
Transfer movements from a reference video to any character image. Pro mode delivers higher quality output, ideal for complex dance moves and gestures.
Kling 2.6 Pro: Top-tier text-to-video with cinematic visuals, fluid motion, and native audio generation.
Transfer movements from a reference video to any character image. Cost-effective mode for motion transfer, perfect for portraits and simple animations.
Kling's Native 4K is a video generation model that directly outputs professional-grade 4K video in one step, eliminating the need for post-production upscaling
Kling's Native 4K is a video generation model that directly outputs professional-grade 4K video in one step, eliminating the need for post-production upscaling
Kling 3.0 Pro: Top-tier image-to-video with cinematic visuals, fluid motion, and native audio generation, with custom element support.
Transfer movements from a reference video to any character image. Cost-effective mode for motion transfer, perfect for portraits and simple animations.
Kling 3.0 Pro: Top-tier text-to-video with cinematic visuals, fluid motion, and native audio generation, with multi-shot support.
Kling 3.0 Standard: Top-tier image-to-video with cinematic visuals, fluid motion, and native audio generation, with custom element support.
Transfer movements from a reference video to any character image. Cost-effective mode for motion transfer, perfect for portraits and simple animations.
Kling 3.0 Standard: Top-tier text-to-video with cinematic visuals, fluid motion, and native audio generation, with multi-shot support.
Generate high quality 1080p videos from images using Kling's Turbo 3.0 model, with improved lipsync and multishot generation capabilities.
Generate high quality 1080p videos using Kling's Turbo 3.0 model, with improved lipsync and multishot generation capabilities.
Kling 3.0 Turbo Standard animates a first and last frame reference image into 720P video with native audio, delivering quick, affordable image-driven motion for fast turnaround
Kling 3.0 Turbo Standard is a fast, cost-efficient video generation model that turns text prompts directly into 720P video with native audio, optimized for rapid iteration and high-volume production
Generate audio from input videos using Kling
Kokoro is a lightweight text-to-speech model that delivers comparable quality to larger models while being significantly faster and more cost-efficient.
A natural and expressive Brazilian Portuguese text-to-speech model optimized for clarity and fluency.
A high-quality British English text-to-speech model offering natural and expressive voice synthesis.
An expressive and natural French text-to-speech model for both European and Canadian French.
A fast and expressive Hindi text-to-speech model with clear pronunciation and accurate intonation.
A high-quality Italian text-to-speech model delivering smooth and expressive speech synthesis.
A fast and natural-sounding Japanese text-to-speech model optimized for smooth pronunciation.
A highly efficient Mandarin Chinese text-to-speech model that captures natural tones and prosody.
A natural-sounding Spanish text-to-speech model optimized for Latin American and European Spanish.
Live transcription over WebSocket, with partial results as the speaker talks.
Speak in one language, hear it in another. Streaming speech-to-speech translation with the translated text alongside the audio.
Accurate transcription of an audio clip in one request, with word timestamps, tuned for Japanese and East Asian languages.
Natural Japanese speech, streamed back as it is generated so playback starts almost immediately.
Train a custom LoRA on your own images to teach Krea 2 a new subject, character, or style. Provide a set of training images (and an optional trigger word), and the trainer outputs LoRA weights you can use for inference with the Krea 2 LoRA endpoint.
Generate high-fidelity images from text in seconds with Krea 2 Turbo, the speed-optimized open-source version of Krea 2, preserving its aesthetic range for rapid ideation.
Generate high-fidelity images from text with Krea 2 using a custom-trained LoRA. Apply your LoRA weights to carry a learned subject, character, or style into new generations, with aspect ratio, creativity, and seed controls.
Generate high-fidelity images from text with Krea 2 using a style reference image. Apply a reference image to guide the visual style into new generations, with aspect ratio, creativity, and seed controls.
Generate high-fidelity images from text with Krea 2 Large, supporting aspect ratio, creativity, seed controls, and optional style references.
Generate high-quality images from text with Krea 2 Medium, supporting aspect ratio, creativity controls, seeds, and optional style references.
Generate high-fidelity images extremely fast from text with Krea 2 Medium Turbo, supporting aspect ratio, creativity, seed controls, and optional style references.
Edits and combines reference images using text instructions, with optional prompt expansion, output-size controls, and reproducible generation through a seed.
Edits an input image in a realtime session using text prompts, with seed control.
Generates images from text prompts with optional prompt expansion, configurable output dimensions, and seed control.
Fast Text-to-Video endpoint for Krea's Wan 14b model.
Superfast video model based on Wan 2.1 14b by Krea, excelling at real-time video-editing.
LatentSync is a video-to-video model that generates lip sync animations from audio using advanced algorithms for high-quality synchronization.
LTX 2.5 Alpha Gen extracts frame-matched foreground mattes from existing video, preserving partial transparency in hair and soft edges. It returns a grayscale matte MP4 and a ProRes 4444 cutout with straight alpha for compositing.
LTX-2.5 is Lightricks' open-source audio-video model. This endpoint generates video timed to a supplied audio clip in a speed-optimized mode — useful for music-driven content, dialogue-led shorts, and ads keyed to a track.
LTX-2.5 is Lightricks' open-source audio-video model. This endpoint generates video timed to a supplied audio clip in a quality-optimized mode, for final visuals synchronized to music, dialogue, or a soundtrack.
LTX 2.5 SDR to HDR converts existing SDR video for HDR finishing workflows. It returns a processed video and a ZIP archive of 16-bit half-float OpenEXR frames, with selectable ACEScg, ACEScct, linear sRGB, or ACES2065-1 color space.
LTX-2.5 is Lightricks' open-source audio-video model. This endpoint animates a still image into video with synchronized audio in a single pass, in a speed-optimized mode for quick iteration.
LTX-2.5 is Lightricks' open-source audio-video model. This endpoint animates a still image into video with synchronized audio in a single pass, in a quality-optimized mode for high-fidelity final output.
LTX 2.5 Reframe adapts existing video to a new aspect ratio by generating missing frame content. It supports five aspect ratios, 720p and 1080p output, and source placement controls for fitting and positioning the original video.
LTX-2.5 is Lightricks' open-source audio-video model. This endpoint generates synchronized video and audio from a text prompt in a single pass, in a speed-optimized mode built for rapid iteration and previews.
LTX-2.5 is Lightricks' open-source audio-video model. This endpoint generates synchronized video and audio from a text prompt in a single pass, in a quality-optimized mode for final, high-fidelity output.
LTX 2.5 Upscale increases the resolution of existing H.264 or HEVC video to 1080p, 4K, or 8K. It preserves the source aspect ratio, retains audio in stereo, and rounds the source frame rate to an integer.
Transfer expression from a video to a portrait.
Transfer expression from a video to a portrait.
Generate long videos from images using LongCat Video Distilled
Generate long videos in 720p/30fps from images using LongCat Video Distilled
Generate long videos from text using LongCat Video Distilled
Generate long videos in 720p/30fps from text using LongCat Video Distilled
Generate long videos from images using LongCat Video
Generate long videos in 720p/30fps from images using LongCat Video
Generate long videos from text using LongCat Video
Generate long videos in 720p/30fps from text using LongCat Video
Run Any Stable Diffusion model with customizable LoRA weights.
Run Any Stable Diffusion model with customizable LoRA weights.
Run Any Stable Diffusion model with customizable LoRA weights.
Generate video with audio from audio, text and images using LTX-2
Generate video with audio from audio, text and images using LTX-2.3 and custom LoRA
Generate video with audio from audio, text and images using LTX-2 Distilled
Generate video with audio from audio, text and images using LTX-2.3 Distilled and custom LoRA
Generate video with audio from images using LTX-2.3 Distilled
Generate video with audio from images using LTX-2.3 Distilled and custom LoRA
Generate video with audio from reference videos using LTX-2.3 Distilled
Generate video with audio from reference videos using LTX-2.3 Distilled and custom LoRA
Generate video with audio from text using LTX-2.3 Distilled
Generate video with audio from text using LTX-2.3 Distilled and custom LoRA
Generate video with audio from videos using LTX-2.3 Distilled
Generate video with audio from videos using LTX-2.3 Distilled and custom LoRA
Extend video with audio using LTX-2.3
Extend video with audio using LTX-2.3 and custom LoRA
Generate video with audio from images using LTX-2.3
Generate video with audio from images using LTX-2.3 and custom LoRA
Generate video with audio from reference video, text and images using LTX-2.3
Generate video with audio from reference video, text and images using LTX-2.3 and custom LoRA
Generate video with audio from text using LTX-2.3
Generate video with audio from text using LTX-2.3 and custom LoRA
Generate video with audio from videos using LTX-2.3
Generate video with audio from videos using LTX-2.3 and custom LoRA
LTX-2.3 is a high-quality, fast AI video model available in Pro and Fast variants for text-to-video, image-to-video, and audio-to-video.
LTX-2.3 is a high-quality, fast AI video model available in Pro and Fast variants for text-to-video, image-to-video, and audio-to-video.
LTX-2.3 is a high-quality, fast AI video model available in Pro and Fast variants for text-to-video, image-to-video, and audio-to-video.
LTX-2.3 is a high-quality, fast AI video model available in Pro and Fast variants for text-to-video, image-to-video, and audio-to-video.
Generate high-quality video with audio from audio, text and images using LTX-2.3
Generate high-quality video with audio from audio, text and images using LTX-2.3 and custom LoRA
Remove character from your video using Ltx 2.3
Colorize high-quality video using LTX-2.3
Cross-eyes for high-quality video using LTX-2.3
Day to Night for high-quality video using LTX-2.3
Deblur high-quality video using LTX-2.3
Decompression / Denoise high-quality video using LTX-2.3
Extend high-quality video with audio from input video using LTX-2.3
Extend high-quality video with audio from input video using LTX-2.3 with Lora
Generate HDR from reference video using LTX-2.3
Generate HDR from reference video using LTX-2.3 with lora
Generate high-quality video with audio from images using LTX-2.3
Generate high-quality video with audio from images using LTX-2.3 and custom LoRA
Generate high-quality video with audio from reference, character sheet, storyboard using LTX-2.3
Inpaint high-quality video using LTX-2.3
Inpaint high-quality video using LTX-2.3 with lora
Instant shave high-quality video using LTX-2.3
Outpaint high-quality video using LTX-2.3
Outpaint high-quality video using LTX-2.3 with Lora
Generate high-quality video with audio from reference video, text and images using LTX-2.3
Generate high-quality video with audio from reference video, text and images using LTX-2.3 and custom LoRA
Transform your 3D video render into realistic using first frame with Ltx 2.3
Text to Audio high-quality using LTX-2.3
Text to Audio high-quality using LTX-2.3 with Lora
Generate high-quality video with audio from text using LTX-2.3
Generate high-quality video with audio from text using LTX-2.3 and custom LoRA
Water Simulation transformation for high-quality video using LTX-2.3
LTX-2.3 Reframe converts your videos to any aspect ratio without destructive cropping. It intelligently recenters the original footage and generatively fills the newly exposed areas with content that seamlessly matches the scene, so the result looks like it was shot natively in the target format. Turn landscape footage into vertical 9:16 for social, square 1:1 for feeds, or anything in between. Supports videos up to 60 seconds, with 720p and 1080p outputs across 1:1, 4:5, 5:4, 9:16 and 16:9.
LTX-2.3 is a high-quality, fast AI video model available in Pro and Fast variants for text-to-video, image-to-video, and audio-to-video.
LTX-2.3 is a high-quality, fast AI video model available in Pro and Fast variants for text-to-video, image-to-video, and audio-to-video.
LTX-2.3 is a high-quality, fast AI video model available in Pro and Fast variants for text-to-video, image-to-video, and audio-to-video.
Generate videos from prompts using LTX Video 0.9.1
Generate videos from prompts using LTX Video-0.9.7 13B Distilled and custom LoRA
Extend videos using LTX Video-0.9.7 13B Distilled and custom LoRA
Generate videos from prompts and images using LTX Video-0.9.7 13B Distilled and custom LoRA
Generate videos from prompts, images, and videos using LTX Video-0.9.7 13B Distilled and custom LoRA
Generate videos from images using LTX Video 0.9.1
Train a LoRA that transforms one audio clip into another, learning a reference→target mapping from paired audio examples.
Train a LoRA that generates video from a start image plus a conditioning audio track, producing motion that matches the sound.
Train a LoRA that continues an audio clip forward in time, generating the audio that follows a short clean prefix.
Train a LoRA that generates the lead-in to an audio clip, extending audio backward in time from its ending.
Train a LoRA that regenerates masked time spans of an audio clip while keeping the rest unchanged.
Train a LoRA for a joint audio+video transformation, conditioned on a reference clip (its video and audio) to produce a matching target clip.
Train a LoRA that regenerates a masked video region (guided by kept pixels and a video reference) while jointly generating audio from an audio reference.
Train a LoRA that continues a video forward in time — supply an opening clip at inference and the model generates what comes next.
Train a LoRA that generates the lead-in to a video, extending a clip backward in time from its ending.
Fine-tune LTX 2.3 to animate a starting image — supply a still plus a prompt at inference and the model generates a video that begins from that frame.
Train an IC-LoRA that transforms one audio clip into another, conditioned at inference on a reference audio clip.
Train an IC-LoRA for a joint audio+video transformation, conditioned on a reference clip's video and audio to produce a matching target.
Train an IC-LoRA that regenerates a masked video region (guided by kept pixels and a video reference) while jointly generating audio from an audio reference.
Train an IC-LoRA that learns a video-to-video transformation from paired before/after clips, conditioned at inference on a reference (control) video.
Train an IC-LoRA that regenerates only the masked region of a video, guided by the kept pixels and a separate reference/control video.
Train a LoRA that regenerates a masked region of a video while keeping the rest unchanged, blending the new content with its surroundings.
Train a LoRA that generates the video between keyframes — supply first/last (and optional middle) frames at inference and the model fills the in-between motion.
Train a LoRA that expands the video frame outward, keeping an inner rectangle fixed and generating the surrounding region.
Train a LoRA that generates audio from a text prompt — the audio counterpart of text-to-video — learning a sound or style from your clips.
Fine-tune LTX 2.3 on your own clips to teach it a new subject, character, object, or visual style, then generate full videos from a text prompt.
Train a LoRA that generates audio (foley / sound design) for a silent video, learning a soundtrack that matches the on-screen action.
Train a LoRA that learns a video-to-video transformation from paired before/after clips, steered at inference by a reference (control) video.
Train a LoRA that regenerates only the masked region of a video, guided by both the kept pixels and a separate reference/control video.
Train LTX-2.3 22B for custom styles and effects.
Generate long videos from prompts using LTX Video-0.9.8 13B Distilled and custom LoRA
Extend videos using LTX Video-0.9.8 13B Distilled and custom LoRA
Generate long videos from prompts and images using LTX Video-0.9.8 13B Distilled and custom LoRA
Generate long videos from prompts, images, and videos using LTX Video-0.9.8 13B Distilled and custom LoRA
Luma Ray 3.2 animates a source image into cinematic motion guided by a text prompt, preserving the starting frame's look while controlling resolution, duration, and seamless looping.
Luma Ray 3.2 reframes an existing video into a new aspect ratio guided by a text prompt, preserving the original footage frame-for-frame while controlling resolution and outpainting the surrounding canvas.
Luma Ray 3.2 generates cinematic video from a text prompt, with control over resolution, duration, and seamless looping, plus reference images to lock in subject and style.
Luma Ray 3.2 re-renders an existing video into new cinematic motion guided by a text prompt, preserving the source's look and movement while controlling resolution, duration, and HDR.
Luma Uni-1 Edit reworks a source image from a text instruction, preserving the original composition while applying style changes and following optional reference images to steer the result.
Luma Uni-1 Max generates a single image at the model's highest fidelity, delivering richer detail and stronger prompt adherence than the base tier for hero-quality stills.
Luma Uni-1 Max Edit applies text-guided edits to a source image at maximum fidelity, holding the original structure while honoring reference images for precise, high-detail revisions.
Luma Uni-1 turns a text prompt into a single high-fidelity image, with control over aspect ratio and visual style, plus optional web-sourced and reference-image guidance for sharper grounding.
Lyria 2 is Google's latest music generation model, you can generate any type of music with this model.
Lyria 3 is most recent music model from Google
Lyria 3 Pro is the latest music model from Google
Estimate depth from a single image with Marigold V2, a diffusion-based depth model built on Qwen-Image-Edit, returning a colorized depth map.
Marlin is a 2B video VLM tuned for the two questions developers actually want to ask of their videos: what is happening, and when?
Marlin is a 2B video VLM tuned for the two questions developers actually want to ask of their videos: what is happening, and when?
Meshy N1 retexture applies new textures to an existing 3D model using native 3D texturing, producing 2K base color textures from reference images.
Meshy Relief converts an image into a 3D relief model with a raised surface and a flat back.
Rig humanoid 3D models from GLB URLs with Meshy, returning rigged GLB/FBX files plus basic animations.
Meshy auto-rigs a humanoid 3D model fitting a skeleton and binding the mesh, then applies several motion presets from its animation library
Meshy T2.5 is Meshy's latest Smart Topology model. It converts a single image into a clean, triangle-only 3D mesh with natively separated parts, with a target polycount of up to 30,000 triangles (default 15,000). Compared with Meshy T2, it produces sharper detail and keeps shape and proportions closer to the input image, at the cost of longer generation times; optional textures, PBR maps and humanoid rigging are supported.
Meshy T2.5 is Meshy's latest Smart Topology model. It generates a clean, triangle-only 3D mesh with natively separated parts from a text prompt, with a target polycount of up to 30,000 triangles (default 15,000). Preview mode returns the untextured mesh, and full mode textures it with Meshy 7.1; compared with Meshy T2, it produces sharper detail at the cost of longer generation times.
Meshy T2 is Meshy's faster Smart Topology model. It converts a single image into a clean, triangle-only 3D mesh with natively separated parts, with a target polycount of up to 15,000 triangles (default 4,000). It generates faster than Meshy T2.5, which suits low-poly assets and speed-sensitive pipelines; optional textures, PBR maps and humanoid rigging are supported.
Meshy T2 is Meshy's faster Smart Topology model. It generates a clean, triangle-only 3D mesh with natively separated parts from a text prompt, with a target polycount of up to 15,000 triangles (default 4,000). Preview mode returns the untextured mesh, and full mode textures it with Meshy 7.1; it generates faster than Meshy T2.5, which suits low-poly assets.
Meshy 6 generates 3D models from a single image. It is an earlier standard model from Meshy; Meshy 7.1 is the latest standard model.
Meshy 6 Lite retexture applies new textures to an existing 3D model from a text prompt or a reference image, with optional PBR maps.
Meshy 6 Preview generates 3D models from a single image. It is an earlier preview model from Meshy; Meshy 7.1 is the latest standard model.
Meshy 6 Preview generates 3D models from text prompts. It is an earlier preview model from Meshy; Meshy 7.1 is the latest standard model.
Meshy 6 retexture applies new textures to an existing 3D model from a text prompt or a reference image, with optional PBR maps.
Meshy 6 generates 3D models from text prompts. It is an earlier standard model from Meshy; Meshy 7.1 is the latest standard model.
Meshy 7.1 is the latest standard model from Meshy. It generates 3D models from a single image, with optional textures and PBR maps.
Meshy 7.1 is the latest standard model from Meshy. It generates 3D models from one to four views of the same object, with optional textures and PBR maps.
Meshy 7.1 is the latest standard model from Meshy. It generates 3D models from text prompts, with untextured preview and textured full modes.
Meshy 7 generates 3D models from a single image, with optional textures and PBR maps. Meshy 7.1 is the latest standard model from Meshy.
Meshy 7 generates 3D models from multiple images of the same object, with optional textures and PBR maps. Meshy 7.1 is the latest standard model from Meshy.
Meshy 7 retexture applies new textures to an existing 3D model from a text prompt or a reference image, with optional PBR maps.
Meshy 7 generates 3D models from text prompts, with optional textures and PBR maps. Meshy 7.1 is the latest standard model from Meshy.
Meta's Muse Image model does precise edits that change only what you ask, stay coherent across turns, and compose from multiple reference images.
Meta's Muse Image model has faithful instruction-following and exceptional visual fidelity, with fine details like text, plots, and QR codes rendered accurately.
MAI-Image-2.5 is Microsoft's photorealistic image generation and editing model that turns text prompts or uploaded images into high-quality, design-ready visuals with fine-grained, pixel-level control.
MAI-Image-2.5 is Microsoft's photorealistic image generation and editing model that turns text prompts or uploaded images into high-quality, design-ready visuals with fine-grained, pixel-level control.
Generate high-fidelity, design-ready images with precise typography, strong prompt alignment, and rich visual detail using Microsoft's flagship MAI Image 2.5 Pro.
Apply precise, controllable edits to a reference image while preserving composition, typography, identity, and fine visual detail.
MAI-Image-2.7 edits images from text instructions and up to five input images, with preset or custom dimensions and PNG, JPEG, or WebP output.
MAI-Image-2.7 Flash edits images from text instructions and up to five input images, using Microsoft’s lower-cost Flash variant with PNG, JPEG, or WebP output.
MAI-Image-2.7 Flash is Microsoft’s lower-cost image generation variant, producing images from text prompts with preset or custom dimensions and PNG, JPEG, or WebP output.
MAI-Image-2.7 is Microsoft’s image model for generating images from text prompts, with preset or custom image dimensions and PNG, JPEG, or WebP output.
Estimates metric 3D geometry from a single image, returning a point map, depth, a validity mask and the camera field of view, with surface normals on the normal-capable checkpoints. Four ViT checkpoints trade accuracy for speed, and the result can be exported as a textured GLB mesh and a colored PLY point cloud.
Estimates metric 3D geometry from a single image, returning a point map, depth, surface normals, a validity mask and the camera field of view. A sparse 3D refiner sharpens the predicted depth, and the result can be exported as a textured GLB mesh and a colored PLY point cloud.
Train a MiniMax H3 LoRA on first/last/both keyframe signatures, teaching it to generate video with audio that starts on one image and lands on another.
Train a MiniMax H3 LoRA with first-frame conditioning, so a still image animates into video with audio; captions optional.
MiniMax H3 Image Edit transforms and combines up to 16 reference images using text instructions, supporting targeted edits, style changes, and multi-image composition. Its unified generation and editing model produces detailed results at custom output sizes up to 20 megapixels.
MiniMax H3 Image Highspeed Edit transforms and combines up to 16 reference images using text instructions. It supports multi-image composition and custom output sizes up to 20 megapixels, preserving the first image’s aspect ratio by default.
MiniMax H3 Image Highspeed generates detailed images from text prompts using the Highspeed variant of MiniMax H3 Image. It supports custom output sizes up to 20 megapixels and optional prompt expansion that develops short prompts into complete image descriptions.
MiniMax H3 Image generates detailed images from text prompts, with precise instruction following across photorealistic scenes, illustrations, and visual designs. It supports native high-resolution generation up to 20 megapixels and optional prompt expansion that develops short prompts into complete image descriptions.
MiniMax H3 is a frontier video model. This endpoint animates a supplied image into 2K video, using it as the opening frame or pairs a first and last frame to control a transition between two images with the aspect ratio following the input.
Images into video with synchronized audio using MiniMax H3; your image becomes the first frame, prompt optional, with trained LoRA support for subject consistency.
Transform Blender renders and 3D previs into photorealistic video. H3 Max uses the source clip to guide scene layout, camera movement, and timing, with optional image references for appearance.
H3 Max Multi Angle turns a single image into a video with precise, keyframe-based control over the camera's orbit, elevation, and distance in 3D space
Direct continuous, realtime video streams with live prompts while preserving characters, settings, and story continuity.
H3 Max Extend Video adds a text-guided continuation to an existing video. It supports prompt expansion, adjustable duration and aspect ratio, and output resolutions from 480p to 2K, returning either the full extended video or only the new footage.
fal's H3 Max is a post-trained variant of MiniMax H3, tuned for stronger prompt adherence and better aesthetics while co-optimized with our custom inference stack for higher throughput with no compromises on output quality
Insert a new scene into an existing video with H3 Max. Guide the scene with a prompt, reference images, or reference videos, then return to the original footage.
H3 Max Lip Sync generates a video from an image and supplied audio, synchronizing mouth movements to the soundtrack. It supports optional transcription guidance and output resolutions from 480p to 2K.
Recast the people in a video using reference photos with H3 Max, while preserving the source motion, camera, cuts, and audio.
Direct a continuous world in real time with image references and text prompts.
fal's H3 Max is a post-trained variant of MiniMax H3, tuned for stronger prompt adherence and better aesthetics while co-optimized with our custom inference stack for higher throughput with no compromises on output quality
Expands existing videos to a selected aspect ratio with generative outpainting, preserving source audio and timing. Supports optional text guidance and 480p, 768p, or 1080p output.
Relight any video with H3 Max from a lighting sphere image, while preserving the source subjects, motion, camera and audio.
fal's H3 Max is a post-trained variant of MiniMax H3, tuned for stronger prompt adherence and better aesthetics while co-optimized with our custom inference stack for higher throughput with no compromises on output quality
Extend an existing video with H3 Max Turbo: add 1 to 15 seconds of prompt-guided footage, with optional audio references and output resolutions from 480p to 2K.
fal's H3 Max Turbo is a post-trained variant of MiniMax H3, tuned for stronger prompt adherence and better aesthetics while co-optimized with our custom inference stack for higher throughput with no compromises on output quality
Generate video with synchronized audio from a prompt and image, video, or audio references with H3 Max Turbo, at 480p, 768p, or 1080p for up to 15 seconds.
fal's H3 Max Turbo is a post-trained variant of MiniMax H3, tuned for stronger prompt adherence and better aesthetics while co-optimized with our custom inference stack for higher throughput with no compromises on output quality
Train a MiniMax H3 LoRA with reference conditioning, so different modalities animate into video with audio; captions optional.
MiniMax H3 is a frontier video model. This endpoint generates 2K video from multimodal references up to 9 images for subject and style, 3 video clips for motion, and 3 audio clips each cited in the prompt by order, keeping subjects consistent while following the referenced motion and audio.
References into video with synchronized audio using MiniMax H3
Train a MiniMax H3 LoRA on your own captioned clips for pure text-to-video generation with matching audio.
MiniMax H3 is a frontier video model. This endpoint generates video from a text prompt alone, rendering at 2K in durations from 5 to 15 seconds across seven aspect ratios.
Generate video with synchronized audio from a text prompt using MiniMax H3; load a trained LoRA at adjustable strength to lock in style, character, or motion.
Create blazing fast and economical videos with MiniMax Hailuo-02 Image To Video API at 512p resolution
MiniMax Hailuo-02 Image To Video API (Pro, 1080p): Advanced image-to-video generation model with 1080p resolution
MiniMax Hailuo-02 Text To Video API (Pro, 1080p): Advanced video generation model with 1080p resolution
MiniMax Hailuo-02 Image To Video API (Standard, 768p, 512p): Advanced image-to-video generation model with 768p and 512p resolutions
MiniMax Hailuo-02 Text To Video API (Standard, 768p): Advanced video generation model with 768p resolution
MiniMax Hailuo-2.3-Fast Image To Video API (Pro, 1080p): Advanced fast image-to-video generation model with 1080p resolution
MiniMax Hailuo-2.3-Fast Image To Video API (Standard, 768p): Advanced fast image-to-video generation model with 768p resolution
MiniMax Hailuo-2.3 Image To Video API (Pro, 1080p): Advanced image-to-video generation model with 1080p resolution
MiniMax Hailuo-2.3 Text To Video API (Pro, 1080p): Advanced text-to-video generation model with 1080p resolution
MiniMax Hailuo-2.3 Image To Video API (Standard, 768p): Advanced image-to-video generation model with 768p resolution
MiniMax Hailuo-2.3 Text To Video API (Standard, 768p): Advanced text-to-video generation model with 768p resolution
Generate high quality images from text prompts using MiniMax Image-01. Longer text prompts will result in better quality images.
Generate images from text and a reference image using MiniMax Image-01 for consistent character appearance.
Generate music from text prompts using the MiniMax model, which leverages advanced AI techniques to create high-quality, diverse musical compositions.
MiniMax Music 3 is a high-performance music generation model for creating complete songs up to five minutes long
Generate music from text prompts using the MiniMax model, which leverages advanced AI techniques to create high-quality, diverse musical compositions.
Generate music from text prompts using the MiniMax Music 2.0 model, which leverages advanced AI techniques to create high-quality, diverse musical compositions.
MiniMax Music 2.5 creates complete tracks with singing, backing music, and detailed arrangements from lyrics and a style description.
MiniMax Music 2.6 creates complete tracks with singing, backing music, and detailed arrangements from lyrics and a style description.
Generate speech from text prompts and different voices using the MiniMax Speech-02 HD model, which leverages advanced AI techniques to create high-quality text-to-speech.
Generate fast speech from text prompts and different voices using the MiniMax Speech-02 Turbo model, which leverages advanced AI techniques to create high-quality text-to-speech.
Generate speech from text prompts and different voices using the MiniMax Speech-02 HD model, which leverages advanced AI techniques to create high-quality text-to-speech.
Generate fast speech from text prompts and different voices using the MiniMax Speech-02 Turbo model, which leverages advanced AI techniques to create high-quality text-to-speech.
Generate speech from text prompts and different voices using the MiniMax Speech-2.6 HD model, which leverages advanced AI techniques to create high-quality text-to-speech.
Generate speech from text prompts and different voices using the MiniMax Speech-2.6 HD model, which leverages advanced AI techniques to create high-quality text-to-speech.
Generate speech from text prompts and different voices using the MiniMax Speech-2.8 HD model, which leverages advanced AI techniques to create high-quality text-to-speech.
Generate speech from text prompts and different voices using the MiniMax Speech-2.8 Turbo model, which leverages advanced AI techniques to create high-quality text-to-speech.
Generate video clips from your prompts using MiniMax model
Generate video clips more accurately with respect to natural language descriptions and using camera movement instructions for shot control.
Generate video clips more accurately with respect to initial image, natural language descriptions, and using camera movement instructions for shot control.
Generate video clips from your images using MiniMax Video model
Generate video clips from your prompts using MiniMax model
Generate video clips from your images using MiniMax Video model
Generate video clips maintaining consistent, realistic facial features and identity across dynamic video content
Clone a voice from a sample audio and generate speech from text prompts using the MiniMax model, which leverages advanced AI techniques to create high-quality text-to-speech.
Design a personalized voice from a text description, and generate speech from text prompts using the MiniMax model, which leverages advanced AI techniques to create high-quality text-to-speech.
Generates sound effects and ambience from a text description, up to 600 seconds long, with up to 4 variations per request.
Generates synchronized sound effects for a video as audio files, with speech preserved or separated. Supports text prompts for off-screen sounds and ambience.
Adds synchronized sound effects to a video and returns the video with the sound mixed in. Speech is preserved by default, with text prompts for off-screen sounds and ambience.
MMAudio generates synchronized audio given video and/or text inputs. It can be combined with video models to get videos with audio.
MMAudio generates synchronized audio given text inputs. It can generate sounds described by a prompt.
MoonDreamNext is a multimodal vision-language model for captioning, gaze detection, bbox detection, point detection, and more.
MoonDreamNext Batch is a multimodal vision-language model for batch captioning.
MoonDreamNext Detection is a multimodal vision-language model for gaze detection, bbox detection, point detection, and more.
Moondream2 is a highly efficient open-source vision language model that combines powerful image understanding capabilities with a remarkably small footprint.
Moondream2 is a highly efficient open-source vision language model that combines powerful image understanding capabilities with a remarkably small footprint.
Moondream2 is a highly efficient open-source vision language model that combines powerful image understanding capabilities with a remarkably small footprint.
Moondream2 is a highly efficient open-source vision language model that combines powerful image understanding capabilities with a remarkably small footprint.
Moondream 3 is a vision language model that brings frontier-level visual reasoning with native object detection, pointing, and OCR capabilities to real-world applications requiring fast, inexpensive inference at scale.
Moondream 3 is a vision language model that brings frontier-level visual reasoning with native object detection, pointing, and OCR capabilities to real-world applications requiring fast, inexpensive inference at scale.
Moondream 3 is a vision language model that brings frontier-level visual reasoning with native object detection, pointing, and OCR capabilities to real-world applications requiring fast, inexpensive inference at scale.
Moondream 3 is a vision language model that brings frontier-level visual reasoning with native object detection, pointing, and OCR capabilities to real-world applications requiring fast, inexpensive inference at scale.
Moondream 3 is a vision language model that brings frontier-level visual reasoning with native object detection, pointing, and OCR capabilities to real-world applications requiring fast, inexpensive inference at scale.
Generate a video starting from an image as the first frame with Marey, a generative video model trained exclusively on fully licensed data.
Pull motion from a reference video and apply it to new subjects or scenes.
Ideal for matching human movement. Your input video determines human poses, gestures, and body movements that will appear in the generated video.
Generate a video from a text prompt with Marey, a generative video model trained exclusively on fully licensed data.
MOSS-Image-SFT is a text-to-image model specializing in Chinese and English text rendering and structured layout control for precise text placement and object arrangement.
Turn a prompt, an image or a reference video into a seamlessly looping GLSL shader, with a page that plays it live.
Turn a prompt, an image or a reference video into an editable GSAP animation: one self-contained HTML page you can play, tweak and ship.
Upload files to Mureka to use them in other endpoints
Writes song lyrics and a title from a text prompt describing a topic, theme, or song concept.
Creates a lyrics video from generated or uploaded audio, with selectable layouts, aspect ratios, backgrounds, and lyric-row or time-range selection.
Generates podcast audio from an ordered sequence of conversation turns, with text and a voice selection for each turn.
Synthesizes speech from text using a built-in voice preset or an uploaded custom voice.
Continues existing song lyrics, accepting section labels alongside the supplied text and returning extended lyrics.
Analyzes a song from an audio URL and returns a written description, identified instruments, genres, and descriptive tags.
Rewrites a selected time region of a generated song or uploaded audio using replacement lyrics.
Analyzes an uploaded song and returns its duration and recognized lyrics with timing information when available.
Separates a song into downloadable audio stems, with available stem types and optional MIDI output determined by the selected separation model.
Transcribes generated or uploaded music into a downloadable archive that can include MusicXML, PDF sheet music, and timestamped lyrics.
Creates a reusable custom voice from a short vocal recording and returns a voice identifier for supported generation workflows.
Generates instrumental music from a text description or uploaded instrumental reference, with support for multiple variations.
Generates songs from a text prompt or supplied lyrics, with musical style, vocal, reference-audio, and melody controls, plus support for multiple variations.
Generates soundtracks from an uploaded image or video, with optional style and mood guidance, audio-range selection, and multiple variations.
Generates vocal or instrumental tracks for an existing song or uploaded audio, with controls for track type, musical style, generation region, and optional vocal lyrics.
Extends a generated song or uploaded audio from a specified timestamp using supplied lyrics, with head or tail extension depending on the selected model.
Remixes a generated song or uploaded audio with Mureka 8, using complete lyrics and a prompt that controls genre, mood, instrumentation, and arrangement. Accepts source audio from 10 to 350 seconds; vocal-only, accompaniment-only, and instrumental tracks are not supported.
Generates instrumental music from a text description or uploaded instrumental reference, with support for multiple variations.
Generates songs from a text prompt or supplied lyrics, with musical style, vocal, reference-audio, and melody controls, plus support for multiple variations.
Generates soundtracks from an uploaded image or video, with optional style and mood guidance, audio-range selection, and multiple variations.
Remixes a generated song or uploaded audio with Mureka 9.5, using complete lyrics and a prompt that controls genre, mood, instrumentation, arrangement, and vocal gender. Accepts source audio from 10 to 480 seconds, including vocal-only, accompaniment-only, and instrumental tracks.
Generates instrumental music from a text description or uploaded instrumental reference, with support for multiple variations.
Generates songs from a text prompt or supplied lyrics, with musical style, vocal, reference-audio, and melody controls, plus support for multiple variations.
Generates soundtracks from an uploaded image or video, with optional style and mood guidance, audio-range selection, and multiple variations.
Use NAFNet to fix issues like blurriness and noise in your images. This model specializes in image restoration and can help enhance the overall quality of your photography.
Use NAFNet to fix issues like blurriness and noise in your images. This model specializes in image restoration and can help enhance the overall quality of your photography.
Google's famous original image generation and editing model
Nano Banana 2 is Google's new state-of-the-art fast image generation and editing model
Nano Banana 2 is Google's new state-of-the-art image generation and editing model
Google's famous original image generation and editing model
Nano Banana Pro is Google's new state-of-the-art image generation and editing model
Nano Banana Pro is Google's new state-of-the-art image generation and editing model
Nucleus-Image is a text-to-image generation model built on a sparse mixture-of-experts (MoE) diffusion transformer architecture.
Chat and reason with NVIDIA Nemotron 3.5 Super, a hybrid Mamba-Transformer mixture-of-experts model with optional step-by-step thinking.
Ask NVIDIA Nemotron 3.5 Super about a video clip: summarize what happens, find moments and reason across frames.
Ask NVIDIA Nemotron 3.5 Super questions about an image: describe scenes, read text and charts, and reason over what it sees.
Open, efficient reasoning model from NVIDIA. 30B A3B hybrid Transformer-Mamba MoE, built for enterprise agentic workflows.
Audio reasoning variant of NVIDIA's Nemotron 3 Nano Omni. 30B A3B hybrid Transformer-Mamba MoE - accepts audio plus a prompt and returns text.
Video reasoning variant of NVIDIA's Nemotron 3 Nano Omni. 30B A3B hybrid Transformer-Mamba MoE - accepts video plus a prompt and returns text.
Vision reasoning variant of NVIDIA's Nemotron 3 Nano Omni. 30B A3B hybrid Transformer-Mamba MoE - accepts an image plus a prompt and returns text.
Removes objects and their visual effects using natural language, replacing them with contextually appropriate content
Removes box-selected objects and their visual effects, seamlessly reconstructing the scene with contextually appropriate content.
Removes mask-selected objects and their visual effects, seamlessly reconstructing the scene with contextually appropriate content.
GPT Image 2, OpenAI's latest image model, is capable of creating extremely detailed images with fine typography.
Precise image editing that changes only what's asked, keeping subject, composition, and background intact, with reference subjects staying recognizable across styles and successive edits.
OpenAI's default image model for most applications. Fast, high-quality generation with natural lighting, rich textures, and support for complex layouts including transparent backgrounds.
Editing built for the tightest control, edits scoped precisely to the instruction, with subject and composition preserved across many rounds of revision.
OpenAI's precision-focused image model, built for premium visual work, extra fidelity on intricate detail, in exchange for longer generation times.
Precise image editing that changes only what's asked, keeping subject, composition, and background intact, with reference subjects staying recognizable across styles and successive edits.
OpenAI's default image model for most applications. Fast, high-quality generation with natural lighting, rich textures, and support for complex layouts including transparent backgrounds.
Editing built for the tightest control, edits scoped precisely to the instruction, with subject and composition preserved across many rounds of revision.
OpenAI's precision-focused image model, built for premium visual work, extra fidelity on intricate detail, in exchange for longer generation times.
GPT Image 2, OpenAI's latest image model, is capable of making fine-grained, detailed edits to images.
FlashVSR+ upscales video by up to 4x with a one-step diffusion model, offering a fast Turbo mode and a higher-quality Base mode. Output reaches about 4K, and the input audio track is preserved by default.
Run any LLM with fal. Access Claude (Anthropic), ChatGPT / GPT-5 / GPT-4o (OpenAI), Gemini (Google), Grok (xAI), DeepSeek, Llama (Meta), Qwen (Alibaba), Mistral, and 200+ more models through a single API. Supports reasoning, structured output, and streaming. Powered by OpenRouter.
Run any audio capable LLM with fal. Process audio files — transcription, analysis, understanding — using Gemini (Google) models. Supports wav, mp3, aiff, aac, ogg, flac, m4a. Powered by OpenRouter.
Run any decision model with fal, powered by OpenRouter.
Run any LLM (Large Language Model) with fal, powered by OpenRouter.
OpenAI-compatible chat completions API. Drop-in replacement for the OpenAI API — use any OpenAI SDK or client to access Claude, Gemini, Grok, DeepSeek, Llama, Qwen, Mistral, and all OpenAI models (GPT-5, GPT-4o, o3) through fal. Powered by OpenRouter.
Generate text embeddings using OpenAI-compatible API. Access embedding models like text-embedding-3-small, text-embedding-3-large (OpenAI), and other embedding models available through OpenRouter. Drop-in replacement for the OpenAI embeddings API. Powered by OpenRouter.
The OpenRouter Responses API with fal, powered by OpenRouter, provides unified access to a wide range of large language models - including GPT, Claude, Gemini, and many others through a single API interface.
Run any video-capable LLM with fal. Analyze, summarize, and understand video files using Gemini (Google) models. Supports mp4, mpeg, mov, webm, and YouTube links. Powered by OpenRouter.
Run any VLM (Video Language Model) with fal, powered by OpenRouter.
Run any Vision Language Model with fal. Analyze and understand images using Claude (Anthropic), GPT-5 / GPT-4o (OpenAI), Gemini (Google), Grok (xAI), Llama (Meta), Qwen, Pixtral (Mistral), and more. Send one or multiple images for captioning, analysis, OCR, or visual Q&A. Powered by OpenRouter.
Orpheus TTS is a state-of-the-art, Llama-based Speech-LLM designed for high-quality, empathetic text-to-speech generation. This model has been finetuned to deliver human-level speech synthesis, achieving exceptional clarity, expressiveness, and real-time performances.
A unified paradigm for audio-video generation
Ovi can generate videos with audio from image and text inputs.
PATINA creates seamless high-resolution normal, roughness, basecolor (albedo), height (displacement) and metalness maps from images
Generate complete seamlessly tiling PBR materials including normal, roughness, basecolor, height and metalness maps up to 8K
Extract seamless tiling textures with PBR attribute maps from images
Isaac-01 is a multimodal vision-language model from Perceptron for various vision language tasks.
OpenAI spec compatible endpoint of Isaac-01 which is a multimodal vision-language model from Perceptron for various vision language tasks.
Phota's model empowers developers, photographers, and creators with personalized photograph generation and editing.
Generate profiles using 30-50 images of a subject with Phota.
Phota's model enables personalized photo editing, preserving identity while erasing distractions seamlessly.
Enhance images while preserving identities with Phota
Pixelcut’s Background Remover enables fast, ultra high-quality removal of backgrounds from images. Perfect for e-commerce and image editing workflows. Powered by advanced AI for clean, perfect cutouts every time.
Turn one product photo into a seamless 5 to 15 second video loop with a locked camera, subtle ambient motion or a full 360° spin.
Pixelcut's Background Remover produces fast, high-quality cutouts built for e-commerce product imagery
Pixelcut's Video Background Remover is an AI segmentation model that erases backgrounds frame by frame, with seamless temporal consistency.
Animate images into cinematic videos with PixVerse C1, supporting 1080p resolution and native audio generation.
Generate character-consistent videos from reference images using PixVerse C1, with subject and background references.
Generate film-grade videos from text prompts with native audio, up to 1080p and 15 seconds, using PixVerse C1.
Create seamless cinematic transitions between two images with PixVerse C1, with native audio and up to 1080p.
PixVerse Extend model is a video extending tool for your videos using with high-quality video extending techniques
PixVerse Extend model is a video extending tool for your videos using with high-quality video extending techniques
Generate realistic lipsync animations from audio using advanced algorithms for high-quality synchronization with PixVerse Lipsync model
PixVerse VibeMV generates music videos from audio, with optional character references and lyric subtitles. It supports visual style presets, custom style references, five aspect ratios, and output at 720p or 1080p.
Add immersive sound effects and background music to your videos using PixVerse sound effects generation
Generate high quality video clips by swapping person, objects and background using Pixverse Swap.
Generate high quality video clips with different effects using PixVerse v3.5
Generate high quality video clips from text and image prompts using PixVerse v3.5
Generate high quality video clips from text and image prompts quickly using PixVerse v3.5 Fast
Generate high quality video clips from text prompts using PixVerse v3.5
Generate high quality video clips quickly from text prompts using PixVerse v3.5 Fast
Create seamless transition between images using PixVerse v3.5
Generate high quality video clips with different effects using PixVerse v4.5
Generate high quality video clips from text and image prompts using PixVerse v4.5
Generate fast high quality video clips from text and image prompts using PixVerse v4.5
Generate high quality video clips from text and image prompts using PixVerse v4.5
Generate high quality and fast video clips from text and image prompts using PixVerse v4.5 fast
Create seamless transition between images using PixVerse v4.5
Generate high quality video clips with different effects using PixVerse v4
Generate high quality video clips from text and image prompts using PixVerse v4
Generate fast high quality video clips from text and image prompts using PixVerse v4
Generate high quality video clips from text and image prompts using PixVerse v4
Generate high quality and fast video clips from text and image prompts using PixVerse v4 fast
Pixverse Effects
Generate high quality video clips from text and image prompts using PixVerse v5.5
Generate high quality video clips from text and image prompts using PixVerse v5.5
Pixverse Transition
Use the latest pixverse v5.6 model to turn your texts and images into amazing videos.
Use the latest pixverse v5.6 model to turn your texts into amazing videos.
Use the latest pixverse v5.6 model to turn your texts and images into amazing videos.
Generate high quality video clips with different effects using PixVerse v5
Generate high quality video clips from text and image prompts using PixVerse v5
Generate high quality video clips from text and image prompts using PixVerse v5
Create seamless transition between images using PixVerse v5
Pixverse's latest v6 Model.
Pixverse's latest V6 Model
Generate videos from images with PixVerse V6 Max, with optional audio and multi-clip generation.
Generate videos from text with PixVerse V6 Max, with optional audio and multi-clip generation.
Pixverse's latest v6 Model.
Pixverse's latest v6 Model.
Post Processing is an endpoint that can enhance images using a variety of techniques including grain, blur, sharpen, and more.
Apply Gaussian or Kuwahara blur effects with adjustable radius and sigma parameters
Create chromatic aberration by shifting red, green, and blue channels horizontally or vertically with customizable shift amounts.
Adjust color temperature, brightness, contrast, saturation, and gamma values for color correction.
Apply various color tints (sepia, red, green, blue, cyan, magenta, yellow, purple, orange, warm, cool, lime, navy, vintage, rose, teal, maroon, peach, lavender, olive) with adjustable strength.
Reduce color saturation using different methods (luminance Rec.709, luminance Rec.601, average, lightness) with adjustable factor.
Blend two images together using smooth linear interpolation with a configurable blend factor.
Apply dodge and burn effects with multiple modes and adjustable intensity.
Apply film grain effect with different styles (modern, analog, kodak, fuji, cinematic, newspaper) and customizable intensity and scale
Apply a parabolic distortion effect with configurable coefficient and vertex position.
Apply sharpening effects with three modes: basic unsharp mask, smart sharpening with edge preservation, and Contrast Adaptive Sharpening (CAS).
Apply solarization effect by inverting pixel values above a threshold
Add a darkening vignette effect around the edges of the image with adjustable strength
Clone your voices using Qwen3-TTS Clone-Voice model with zero shot cloning capabilities and use it on text-to-speech models to create speeches of yours!
Clone your voices using Qwen3-TTS Clone-Voice model with zero shot cloning capabilities and use it on text-to-speech models to create speeches of yours!
Bring speech to your texts using Qwen3-TTS Custom-Voice model with pre-trained voices or use your custom voice with Qwen3-TTS Clone Voice model
Bring speech to your texts using Qwen3-TTS Custom-Voice model with pre-trained voices or use your custom voice with Qwen3-TTS Clone Voice model
Create custom voices using Qwen3-TTS Voice Design model and later use Clone Voice model to create your own voices!
Qwen-Image is an image generation foundation model in the Qwen series that achieves significant advances in complex text rendering and precise image editing.
Qwen Image 2512 is an improved version of Qwen Image with better text rendering, finer natural textures, and more realistic human generation.
LoRA inference endpoint for Qwen Image 2512, an improved version of Qwen Image with better text rendering, finer natural textures, and more realistic human generation.
Endpoint for Qwen's Image Editing Plus model also known as Qwen-Image-Edit-2509. Has superior text editing capabilities and multi-image support.
LoRA endpoint for the Qwen Image Edit 2509 model.
Add a realistic scene behind the object with white background
Generate full portrait from a cropped face photo
Create group photos
Blend products into backgrounds with automatic perspective and lighting correction
Removes harsh shadows and light spots from images, replacing them with soft, even, natural-looking illumination.
Precise camera position and angle control (rotation, zoom, vertical movement)
Create cinematic transitions and scene progressions (camera movements, framing changes)
Remove unwanted elements (objects, people, text) while maintaining image consistency
Remove existing lighting and apply soft, even illumination
Apply designs/graphics onto people's shirts
Endpoint for Qwen's Image Editing 2511 model.
Endpoint for Qwen's Image Editing 2511 model with LoRa support.
Generates same scene from different angles (azimuth/elevation) with Qwen image Edit 2511 and the Lora Multiple Angles
Image to Image Endpoint for Qwen's Image Editing model. Has superior text editing capabilities.
Inpainting Endpoint for the Qwen Edit Image editing model.
Endpoint for Qwen's Image Editing Plus model also known as Qwen-Image-Edit-2509. Has superior text editing capabilities and multi-image support.
LoRA endpoint for the Qwen Image Edit Plus model.
Add a realistic scene behind the object with white background
Generate full portrait from a cropped face photo
Create group photos
Blend products into backgrounds with automatic perspective and lighting correction
Removes harsh shadows and light spots from images, replacing them with soft, even, natural-looking illumination.
Precise camera position and angle control (rotation, zoom, vertical movement)
Create cinematic transitions and scene progressions (camera movements, framing changes)
Remove unwanted elements (objects, people, text) while maintaining image consistency
Remove existing lighting and apply soft, even illumination
Apply designs/graphics onto people's shirts
Qwen-Image (Image-to-Image) transforms and edits input images with high fidelity, enabling precise style transfer, enhancement, and creative modification.
Qwen-Image-Layered is a model capable of decomposing an image into multiple RGBA layers.
Qwen-Image-Layered is a model capable of decomposing an image into multiple RGBA layers. Use loras to get your custom outputs.
Recraft 20b is a new and affordable text-to-image model.
Enhances a given raster image using the 'creative upscale' tool, increasing image resolution, making the image sharper and cleaner.
Enhances a given raster image using 'crisp upscale' tool, boosting resolution with a focus on refining small details and faces.
Recraft V3 Create Style is capable of creating unique styles for Recraft V3 based on your images.
Recraft V3 is a text-to-image model with the ability to generate long texts, vector art, images in brand style, and much more. As of today, it is SOTA in image generation, proven by Hugging Face's industry-leading Text-to-Image Benchmark by Artificial Analysis.
Recraft V3 is a text-to-image model with the ability to generate long texts, vector art, images in brand style, and much more. As of today, it is SOTA in image generation, proven by Hugging Face's industry-leading Text-to-Image Benchmark by Artificial Analysis.
Recraft V4.1 Flash generates raster images from text prompts, including photography, illustrations, and mixed-media compositions, with controls for image size, color palette, and background color.
Recraft V4.1 Pro pushes the V4.1 model into high-resolution territory — up to 2048×2048 and ultra-wide formats. Made for hero imagery, campaign work, and print, it preserves the same design taste at sizes ready for the final deliverable.
Recraft V4.1 Pro Vector generates large-format, fully editable SVGs with the structural clarity professional illustrators expect. Built for poster art, complex brand assets, and detailed scene illustration, it scales without losing geometric integrity.
Recraft V4.1 builds on the design-first foundation of V4 with sharper prompt control and cleaner composition. Tuned for brand systems and editorial work, it delivers production-ready raster images that hold up next to a designer's hand.
Recraft V4.1 Vector turns prompts into fully editable SVGs with structured layers and clean geometry. Built for logos, icons, and illustration systems, it produces artwork that goes straight from generation into Figma or Illustrator.
Recraft V4.1 Utility Pro pairs the high-resolution output of V4.1 Pro with a faster, cost-efficient runtime. Designed for studios shipping large-format work at scale, it makes premium-quality raster generation viable across full creative pipelines.
Recraft V4.1 Utility is a faster, lighter variant of V4.1 made for high-volume creative workflows. Ideal for ideation, A/B exploration, and content pipelines, it keeps Recraft's design sensibility while optimizing for throughput and cost.
Creates a reusable style from your reference images and returns a style ID you can pass to Recraft V4 Styles image and vector generation.
Creates a reusable style from your reference images for use with Recraft V4 Styles Pro generation
Recraft V4 was developed with designers to bring true visual taste to AI image generation. Built for brand systems and production-ready workflows, it goes beyond prompt accuracy — delivering stronger composition, refined lighting, realistic materials, and a cohesive aesthetic. The result is imagery shaped by professional design judgment, ready for immediate real-world use without additional post-processing.
Recraft V4 was developed with designers to bring true visual taste to AI image generation. Built for brand systems and production-ready workflows, it goes beyond prompt accuracy — delivering stronger composition, refined lighting, realistic materials, and a cohesive aesthetic. The result is imagery shaped by professional design judgment, ready for immediate real-world use without additional post-processing.
Generates raster images that hold a consistent style, from either a saved style ID or reference images attached directly.
Generates vector images that hold a consistent style, from either a saved style ID or reference images attached directly.
Generates raster images that hold a consistent style, from either a saved style ID or reference images attached directly.
Generates vector images that hold a consistent style, from either a saved style ID or reference images attached directly.
Recraft V4 was developed with designers to bring true visual taste to AI image generation. Built for brand systems and production-ready workflows, it goes beyond prompt accuracy delivering stronger composition, refined lighting, realistic materials, and a cohesive aesthetic. The result is imagery shaped by professional design judgment, ready for immediate real-world use without additional post-processing.
Recraft V4 was developed with designers to bring true visual taste to AI image generation. Built for brand systems and production-ready workflows, it goes beyond prompt accuracy — delivering stronger composition, refined lighting, realistic materials, and a cohesive aesthetic. The result is imagery shaped by professional design judgment, ready for immediate real-world use without additional post-processing.
Converts a given raster image to SVG format using Recraft model.
Transform voices using Resemble AI's Chatterbox. Convert audio to new voices or your own samples, with expressive results and built-in perceptual watermarking.
Generate expressive, natural speech with Resemble AI's Chatterbox. Features unique emotion control, instant voice cloning from short audio, and built-in watermarking.
Interpolate images with RIFE - Real-Time Intermediate Flow Estimation
Interpolate videos with RIFE - Real-Time Intermediate Flow Estimation
Explore interactive worlds in real time with movement and camera controls.
Juggernaut Base Flux by RunDiffusion is a drop-in replacement for Flux [Dev] that delivers sharper details, richer colors, and enhanced realism, while instantly boosting LoRAs and LyCORIS with full compatibility.
Juggernaut Base Flux by RunDiffusion is a drop-in replacement for Flux [Dev] that delivers sharper details, richer colors, and enhanced realism, while instantly boosting LoRAs and LyCORIS with full compatibility.
Juggernaut Lightning Flux by RunDiffusion provides blazing-fast, high-quality images rendered at five times the speed of Flux. Perfect for mood boards and mass ideation, this model excels in both realism and prompt adherence.
Juggernaut Pro Flux by RunDiffusion is the flagship Juggernaut model rivaling some of the most advanced image models available, often surpassing them in realism. It combines Juggernaut Base with RunDiffusion Photo and features enhancements like reduced background blurriness.
Juggernaut Pro Flux by RunDiffusion is the flagship Juggernaut model rivaling some of the most advanced image models available, often surpassing them in realism. It combines Juggernaut Base with RunDiffusion Photo and features enhancements like reduced background blurriness.
RunDiffusion Photo Flux provides insane realism. With this enhancer, textures and skin details burst to life, turning your favorite prompts into vivid, lifelike creations. Recommended to keep it at 0.65 to 0.80 weight. Supports resolutions up to 1536x1536.
Sa2VA is an MLLM capable of question answering, visual prompt understanding, and dense object segmentation at both image and video levels
Sa2VA is an MLLM capable of question answering, visual prompt understanding, and dense object segmentation at both image and video levels
Sa2VA is an MLLM capable of question answering, visual prompt understanding, and dense object segmentation at both image and video levels
Sa2VA is an MLLM capable of question answering, visual prompt understanding, and dense object segmentation at both image and video levels
SAM 3.1 builds comes with Object Multiplex, a shared-memory approach for joint multi-object tracking that delivers faster speeds with larger number of objects tracked.
SAM 3.1 builds comes with Object Multiplex, a shared-memory approach for joint multi-object tracking that delivers faster speeds with larger number of objects tracked.
SAM 3.1 builds comes with Object Multiplex, a shared-memory approach for joint multi-object tracking that delivers faster speeds with larger number of objects tracked.
SAM 3.1 builds comes with Object Multiplex, a shared-memory approach for joint multi-object tracking that delivers faster speeds with larger number of objects tracked.
SAM 3D enables full scene reconstructions, placing objects and humans in a shared context together.
SAM 3D allows for accurate 3D reconstruction of human body shape and position from a single image.
SAM 3D enables precise 3D reconstruction of objects from real images, while accurately reconstructing their geometry and texture.
SAM 3 is a unified foundation model for promptable segmentation in images and videos. It can detect, segment, and track objects using text or visual prompts such as points, boxes, and masks.
SAM 3 is a unified foundation model for promptable segmentation in images and videos. It can detect, segment, and track objects using text or visual prompts such as points, boxes, and masks.
SAM 3 is a unified foundation model for promptable segmentation in images and videos. It can detect, segment, and track objects using text or visual prompts such as points, boxes, and masks.
SAM 3 is a unified foundation model for promptable segmentation in images and videos. It can detect, segment, and track objects using text or visual prompts such as points, boxes, and masks.
SAM 3 is a unified foundation model for promptable segmentation in images and videos. It can detect, segment, and track objects using text or visual prompts such as points, boxes, and masks.
Audio separation with SAM Audio. Isolate any sound using natural language—professional-grade audio editing made simple for creators, researchers, and accessibility applications.
Audio separation with SAM Audio. Isolate any sound using natural language—professional-grade audio editing made simple for creators, researchers, and accessibility applications.
Audio separation with SAM Audio. Isolate any sound using natural language—professional-grade audio editing made simple for creators, researchers, and accessibility applications.
SAM 2 is a model for segmenting images automatically. It can return individual masks or a single mask for the entire image.
SAM 2 is a model for segmenting images and videos in real-time.
SAM 2 is a model for segmenting images and videos in real-time.
Sana can synthesize high-resolution, high-quality images with strong text-image alignment at a remarkably fast speed, with the ability to generate 4K images in less than a second.
Sana Sprint is a text-to-image model capable of generating 4K images with exceptional speed.
Sana v1.5 1.6B is a lightweight text-to-image model that delivers 4K image generation with impressive efficiency.
Sana v1.5 4.8B is a powerful text-to-image model that generates ultra-high quality 4K images with remarkable detail.
Use SeedVR2 to upscale your images
Use SeedVR2 to upscale images, retaining seamless tiling
Upscale your videos using SeedVR2 with temporal consistency!
Smart image resize to arbitrary dimensions, powered by Nano Banana Pro with vision-LLM-guided prompting for composition-aware recomposition. Crop, cropping, resize ads.
Rembg-enhance is optimized for 2D vector images, 3D graphics, and photos by leveraging matting technology.
Generates licensed, commercial-use-safe music from a single text prompt, with full control over style, mood, instrumentation, and exact duration.
Generates high-quality, commercial-use-safe sound effects from a text prompt, with full control over type, texture, intensity, and exact duration.
Analyzes your video’s pacing, mood, and timing to generate a frame-synced, licensed, commercial-use-safe soundtrack in seconds.
Analyzes a video and generates synchronized, royalty-free sound effects timed to visible actions. Returns the generated sound-effects audio track for commercial use.
Generates perfectly synced music for any video. Return a licensed music soundtrack ready for commercial use (optional preservation of the original speech in video)
Adds synchronized, royalty-free, commercial-use-safe sound effects to a video. Returns the finished video with the generated audio mixed in.
Leverage the rapid processing capabilities of AI models to enable accurate and efficient real-time speech-to-text transcription.
Leverage the rapid processing capabilities of AI models to enable accurate and efficient real-time speech-to-text transcription.
Leverage the rapid processing capabilities of AI models to enable accurate and efficient real-time speech-to-text transcription.
Leverage the rapid processing capabilities of AI models to enable accurate and efficient real-time speech-to-text transcription.
Idol by Sperid Labs turns a single image of an object into a 3D Gaussian splat, with a GLB point-cloud preview.
Turn one image into a game-ready 3D mesh with clean geometry and a sharp repainted texture, with an optional rig and animations.
Turn one image into a game-ready 3D mesh in about 20 seconds, with an optional rig and animations.
Estimates relative depth from a single image with Iris-Z, a FLUX.2-based joint RGB and depth model, returning a colorized depth map and a float32 log-depth array at the input resolution. An optional multi-seed ensemble combines up to 10 predictions for higher accuracy.
Generates an image that follows the geometry of a depth map with Iris-Z, a FLUX.2-based joint RGB and depth model, guided by an optional text prompt. Accepts depth, disparity or colorized depth maps, keeps the input aspect ratio and renders at 512 or 1024 resolution.
Estimates relative depth from a single image with Iris-Z Small, the lighter variant of the FLUX.2-based joint RGB and depth model, returning a colorized depth map and a float32 log-depth array at the input resolution. An optional multi-seed ensemble combines up to 10 predictions for higher accuracy.
Generates an image that follows the geometry of a depth map with Iris-Z Small, the lighter variant of the FLUX.2-based joint RGB and depth model, at roughly a third of the generation time of Iris-Z. Accepts depth, disparity or colorized depth maps with an optional text prompt, keeps the input aspect ratio and renders at 512 or 1024 resolution.
Generates an image from a text prompt with Iris-Z Small, the lighter variant of the FLUX.2-based joint RGB and depth model, at roughly a third of the generation time of Iris-Z. Supports preset or custom sizes up to 1.31 MP with both sides in multiples of 16, adjustable steps and guidance, and PNG or JPEG output.
Generates an image and its aligned depth together from a text prompt with Iris-Z Small, the lighter variant of the FLUX.2-based joint RGB and depth model, at roughly a third of the generation time of Iris-Z. Returns the RGB image, a colorized depth map and a float32 log-depth array, with sizes up to 1.31 MP.
Generates an image from a text prompt with Iris-Z, a FLUX.2-based joint RGB and depth model. Supports preset or custom sizes up to 1.31 MP with both sides in multiples of 16, adjustable steps and guidance, and PNG or JPEG output.
Generates an image and its aligned depth together from a text prompt with Iris-Z, a FLUX.2-based joint RGB and depth model, returning the RGB image, a colorized depth map and a float32 log-depth array. Supports preset or custom sizes up to 1.31 MP with adjustable steps and guidance.
Open source text-to-audio model.
Generate high quality music and sound effects using Stable Audio 2.5 from StabilityAI
Generate high quality music and sound effects using Stable Audio 2.5 from StabilityAI
Generate high quality music and sound effects using Stable Audio 2.5 from StabilityAI
Stable Audio 3 Medium audio inpainting is a 1.4 billion parameter latent diffusion model that fills in or reworks selected segments of a stereo track guided by text prompts, supporting single- and multi-segment editing.
Stable Audio 3 Medium audio outpainting is a 1.4 billion parameter latent diffusion model that extends existing stereo audio beyond its original endpoint via causal continuation guided by text prompts.
Stable Audio 3 Medium audio-to-audio is a 1.4 billion parameter latent diffusion model that transforms an input audio clip into new stereo variations up to 6 minutes guided by a text prompt.
Stable Audio 3 Medium Base audio inpainting is the foundational 1.4 billion parameter checkpoint for editing or filling selected stereo audio segments guided by text prompts.
Stable Audio 3 Medium Base audio outpainting is the foundational 1.4 billion parameter checkpoint that extends existing stereo audio with causal continuation guided by text prompts.
Stable Audio 3 Medium Base audio-to-audio is the foundational 1.4 billion parameter checkpoint that transforms input audio into new stereo variations up to 6 minutes guided by text prompts.
Stable Audio 3 Medium Base is the foundational 1.4 billion parameter text-to-audio checkpoint generating stereo music up to 6 minutes, intended as the unmodified base for custom fine-tuning workflows.
Stable Audio 3 Medium is a 1.4 billion parameter latent diffusion model that generates high-quality stereo music up to 6 minutes from text prompts, trained on fully licensed data for safe commercial use.
Stable Audio 3 Small Music audio inpainting is a 459 million parameter latent diffusion model that fills in or reworks selected segments of a music track guided by text prompts.
Stable Audio 3 Small Music audio outpainting is a 459 million parameter latent diffusion model that extends music compositions beyond their original endpoint via causal continuation.
Stable Audio 3 Small Music audio-to-audio is a 459 million parameter latent diffusion model that transforms input music into new variations up to 2 minutes guided by text prompts.
Stable Audio 3 Small Music Base audio inpainting is the foundational 459 million parameter checkpoint for editing or filling selected music segments guided by text prompts.
Stable Audio 3 Small Music Base audio outpainting is the foundational 459 million parameter checkpoint that extends music tracks via causal continuation guided by text prompts.
Stable Audio 3 Small Music Base audio-to-audio is the foundational 459 million parameter checkpoint that transforms input music into new variations up to 2 minutes guided by text prompts.
Stable Audio 3 Small Music Base is the foundational 459 million parameter checkpoint generating full music compositions up to 2 minutes from text prompts, intended as the unmodified base for fine-tuning.
Stable Audio 3 Small Music is a 459 million parameter latent diffusion model that generates full stereo music compositions up to 2 minutes from text prompts, lightweight enough for on-device deployment.
Stable Audio 3 Small SFX audio inpainting is a 459 million parameter latent diffusion model that fills in or reworks selected segments of a sound-effect track guided by text prompts.
Stable Audio 3 Small SFX audio outpainting is a 459 million parameter latent diffusion model that extends sound-effect tracks beyond their original endpoint via causal continuation.
Stable Audio 3 Small SFX audio-to-audio is a 459 million parameter latent diffusion model that transforms input audio into new sound-effect variations guided by text prompts.
Stable Audio 3 Small SFX Base audio inpainting is the foundational 459 million parameter checkpoint for editing or filling selected sound-effect segments guided by text prompts.
Stable Audio 3 Small SFX Base audio outpainting is the foundational 459 million parameter checkpoint that extends sound-effect tracks via causal continuation guided by text prompts.
Stable Audio 3 Small SFX Base audio-to-audio is the foundational 459 million parameter checkpoint that transforms input audio into new sound-effect variations guided by text prompts.
Stable Audio 3 Small SFX Base is the foundational 459 million parameter checkpoint generating sound effects from text prompts, intended as the unmodified base for fine-tuning.
Stable Audio 3 Small SFX is a 459 million parameter latent diffusion model that generates high-quality sound effects from text prompts, designed for on-device deployment on mobile phones and consumer laptops.
Stable Diffusion 3.5 Large is a Multimodal Diffusion Transformer (MMDiT) text-to-image model that features improved performance in image quality, typography, complex prompt understanding, and resource-efficiency.
Generate realistic lipsync animations from audio using advanced algorithms for high-quality synchronization.
Use React-1 from SyncLabs to refine human emotions and do realistic lip-sync without losing details!
Generate realistic lipsync animations from audio using advanced algorithms for high-quality synchronization with Sync Lipsync 2.0 model
Generate high-quality realistic lipsync animations from audio while preserving unique details like natural teeth and unique facial features using the state-of-the-art Sync Lipsync 2 Pro model.
sync-3 most powerful lipsync model yet, featuring native visual intelligence for professional-quality video.
sync-3 image to video turns a single still into a talking character, and works with any illustration or animated frame paired with a voice track
TBC mischief 1 is the world's first neuron-derived video model. Generate videos 5x faster at a fraction of the cost, so you can explore more ideas, experiment freely, and make more mischief.
Professional color and lighting correction powered by Topaz Labs. Adjust V2 fixes exposure, White Balance corrects color casts, Colorize adds color to black-and-white photos. Best for one-click photo correction.
Professional motion deblur powered by Topaz Labs. Themis 2 restores clarity to fast-moving, motion-blurred footage at source resolution. Best for sports and action footage.
Professional photo denoising powered by Topaz Labs. Normal, Strong and Extreme presets clean noise at source resolution; Denoise Max adds generative detail recovery. Best for high-ISO and night photography.
Professional video denoising powered by Topaz Labs. Nyx models remove noise at source resolution, with Nyx Fast as a lighter, cheaper pass. Best for low-light and high-ISO footage.
Professional frame interpolation powered by Topaz Labs. Apollo, Chronos and Aion retime footage up to 120 fps, from smooth motion to extreme slow motion. Best for fluid 60fps output and slow-motion effects.
Professional image restoration powered by Topaz Labs. Recover 3 generatively rebuilds natural detail; Dust-Scratch V2 cleans film dust and scratches. Best for old, damaged or degraded photos.
Professional SDR-to-HDR conversion powered by Topaz Labs. Hyperion 2.5 redistributes luminance and color while preserving detail in text, faces and motion. Best for giving flat SDR footage a true HDR look.
Professional photo sharpening powered by Topaz Labs. Models tuned per blur type (lens, motion, portrait, wildlife), plus Super Focus for generative recovery of severely blurred shots. Best for out-of-focus and motion-blurred photos.
Professional creative image upscaling powered by Topaz Labs. Bloom 2 reinvents detail with adjustable creativity and color preservation. Best for AI-generated images that need striking enhancement.
Professional generative image upscaling powered by Topaz Labs. Wonder 3.5 leads the range, with Redefine for prompt-guided detail and Recovery for extreme low-resolution sources. Best for rebuilding sharp detail in small or blurry images.
Professional photo upscaling powered by Topaz Labs. Gigapixel precision models (Standard V2, High Fidelity, Low Resolution, CGI, Text Refine) enlarge images faithfully up to 4x. Best for photos that must stay true to the original.
Professional transparent-image upscaling powered by Topaz Labs. Preserves the alpha channel end to end with PNG output. Best for logos, stickers and assets with transparency.
Professional creative video upscaling powered by Topaz Labs. Astra 2 reimagines fine detail and typically delivers 4K output. Best for cinematic shots that need maximum visual impact.
Professional generative video upscaling powered by Topaz Labs. Starlight models rebuild detail that is not in the source, with Fast variants at half the price. Best for low-quality, compressed or archive footage.
Professional video upscaling powered by Topaz Labs. Precision models (Proteus, Artemis, Iris, Dione, Theia, Gaia, Rhea) enhance footage up to 4x while staying faithful to the source. Best for clean, natural upscales of real-world footage.
Generate 3D models from your images using Trellis. A native 3D generative model enabling versatile and high-quality 3D asset creation.
Generate 3D models from your images using Trellis 2. A native 3D generative model enabling versatile and high-quality 3D asset creation.
Retexture existing 3D models using images with Trellis 2
Generate 3D models from multiple images using Trellis. A native 3D generative model enabling versatile and high-quality 3D asset creation.
Generate high-quality 3D models from a single image using Tripo H3.1.
Generate 3D models from multiple view images using Tripo H3.1.
Generate 3D models from text descriptions using Tripo H3.1.
Generate 3D models from a single image using Tripo P1.
Generate 3D models from text descriptions using Tripo P1.
Tripo P2 generates 3D models from a single image, with optional PBR textures, adjustable face counts, and triangle or quad mesh topology.
Tripo P2 generates 3D models from a text prompt, with optional PBR textures, adjustable face counts, and triangle or quad mesh topology.
Converts triangle meshes into clean quad topology at a target polygon count, animation-ready with no manual retopology.
Automatically splits a 3D model into semantic parts for editing, texturing, and rigging.
State of the art Image to 3D Object generation. Generate 3D model from a single image!
State of the art Multiview to 3D Object generation. Generate 3D models from multiple images!
Real-time games rendered using VAST's Eden Live 14B
Real-time generative rendering from any reference using VAST's Eden Live 14B
Real-time games rendered using VAST's Eden Live 5B
Real-time generative rendering from any reference using VAST's Eden Live 5B
Generate high-quality videos with UGC-like avatars from audio
Generate high-quality videos with UGC-like avatars from text
Studio-quality speech from noisy recordings
Dubs videos up to 10 minutes into 86 languages with cloned voices and lip-sync, preserving each speaker’s tone and emotion.
Dubs videos up to 10 minutes into 94 languages and 14 regional accents with cloned voices and lip-sync. Supports target accent and voice cloning strength controls, multiple speakers, and background audio preservation.
VEED Fabric 1.0 is an image-to-video API that turns any image into a talking video
VEED Fabric 1.0 is an image-to-video API that turns any image into a talking video
VEED Fabric 1.0 text-to-video API
Generate realistic lipsync from any audio using VEED's model.
Generate production-quality lipsync from any audio using VEED's most advanced model yet.
VEED’s Subtitles API transforms raw footage into polished, publish-ready content with professional burned-in subtitles starting at a base rate of $0.10 per minute.
Remove background from any video with people and objects. No green screen needed.
Remove background from any video with people and objects. No green screen needed.
Remove background from videos filmed using chromakey, with automatic green spill suppression for clean, professional edges.
Veo 3 by Google, the most advanced AI video generation model in the world. With sound on!
Veo 3.1 by Google, the most advanced AI video generation model in the world. With sound on!
Extend Veo-Created Videos up to 30 seconds
Faster and more cost effective version of Google's Veo 3.1!
Extend Veo-Created Videos up to 30 seconds
Generate videos from a first/last frame using Google's Veo 3.1 Fast
Generate videos from your image prompts using Veo 3.1 fast.
Generate videos from reference images using Google's Veo 3.1 Fast
Generate videos from a first and last framed using Google's Veo 3.1
Veo 3.1 is the latest state-of-the art video generation model from Google DeepMind
Veo 3.1 Lite balances practical utility with professional capabilities, supporting Text-to-Video and Image-to-Video
Veo 3.1 Lite balances practical utility with professional capabilities, supporting Text-to-Video and Image-to-Video
Veo 3.1 Lite balances practical utility with professional capabilities, supporting Text-to-Video and Image-to-Video
Generate Videos from images using Google's Veo 3.1
Faster and more cost effective version of Google's Veo 3!
Now with a 50% price drop. Generate videos from your image prompts using Veo 3 fast.
Veo 3 is the latest state-of-the art video generation model from Google DeepMind
Turn images or video into a detailed 3D scene with depth, camera poses, and a colored point cloud.
Generate long, expressive multi-voice speech using Microsoft's powerful TTS
Generate long speech snippets fast using Microsoft's powerful TTS.
Generate long, expressive multi-voice speech using Microsoft's powerful TTS
A video understanding model to analyze video content and answer questions about what's happening in the video based on user prompts.
The video upscaler endpoint uses RealESRGAN on each frame of the input video to upscale the video to a higher resolution.
Vidu Image to Video generates high-quality videos with exceptional visual quality and motion diversity from a single image
Vidu Q1 Image to Video generates high-quality 1080p videos with exceptional visual quality and motion diversity from a single image
Generate video clips from your multiple image references using Vidu Q1
Vidu Q1 Start-End to Video generates smooth transition 1080p videos between specified start and end images.
Vidu Q1 Text to Video generates high-quality 1080p videos with exceptional visual quality and motion diversity
Use the latest Vidu Q2 models which much more better quality and control on your videos.
Use the latest Vidu Q2 models which much more better quality and control on your videos.
Vidu Reference-to-Image creates images by using a reference images and combining them with a prompt.
Use the latest Vidu Q2 Pro models which much more better quality and control on your videos.
Use vidu Text-to-Image to turn your prompts into reality.
Use the latest Vidu Q2 models which much more better quality and control on your videos.
Use the latest Vidu Q2 models which much more better quality and control on your videos.
Vidu's latest Q3 pro models.
Vidu's Q3 Turbo Model
Vidu's latest Q3 Reference to Video Mix model
Vidu's latest Q3 pro models
Vidu's Q3 Turbo Model.
Vidu Q4 turns a first-frame image into 3 to 16 second videos with native audio at up to 4K. Best for animating stills, product shots and character scenes with dialogue.
Vidu Q4 generates videos from up to 12 reference images and 3 voice clips, keeping characters, objects, and voices consistent across up to 16 seconds at up to 4K.
Vidu Reference-to-Image creates images by using a reference images and combining them with a prompt.
Vidu Reference to Video creates videos by using a reference images and combining them with a prompt.
Vidu Start-End to Video generates smooth transition videos between specified start and end images.
Vidu Template to Video lets you create different effects by applying motion templates to your images.
VOID removes objects from videos along with all interactions they induce on the scene
Train custom LoRAs for Wan-2.2 T2V/I2V 480P
Train custom LoRAs for Wan-2.2 T2V/I2V 480P
VACE Fun for Wan 2.2 A14B from Alibaba-PAI
VACE Fun for Wan 2.2 A14B from Alibaba-PAI
VACE Fun for Wan 2.2 A14B from Alibaba-PAI
VACE Fun for Wan 2.2 A14B from Alibaba-PAI
Wan 2.5 image-to-image model.
Wan 2.5 image-to-video model.
Wan 2.5 text-to-image model.
Wan 2.5 text-to-video model.
Wan-2.1 is a image-to-video model that generates high-quality videos with high visual quality and motion diversity from images
Add custom LoRAs to Wan-2.1 is a image-to-video model that generates high-quality videos with high visual quality and motion diversity from images
Wan-2.1 Pro is a premium image-to-video model that generates high-quality 1080p videos at 30fps with up to 6 seconds duration, delivering exceptional visual quality and motion diversity from images
Wan-2.1 Pro is a premium text-to-video model that generates high-quality 1080p videos at 30fps with up to 6 seconds duration, delivering exceptional visual quality and motion diversity from text prompts
Add custom LoRAs to Wan-2.1 is a text-to-video model that generates high-quality videos with high visual quality and motion diversity from images
Train custom LoRAs for Wan-2.1 I2V 720P
Train custom LoRAs for Wan-2.1 T2V 1.3B
Train custom LoRAs for Wan-2.1 T2V 14B
Wan-Animate is a video model that generates high-fidelity character videos by replicating the expressions and movements of characters from reference videos.
Wan-Animate Replace is a model that can integrate animated characters into reference videos, replacing the original character while preserving the scene’s lighting and color tone for seamless environmental integration.
Wan-S2V is a video model that generates high-quality videos from static images and audio, with realistic facial expressions, body movements, and professional camera work for film and television applications
Wan 2.2's 5B model produces up to 5 seconds of video 720p at 24FPS with fluid motion and powerful prompt understanding
Wan 2.2's 5B model generates high-resolution, photorealistic images with powerful prompt understanding and fine-grained visual detail
Wan 2.2's 5B model produces up to 5 seconds of video 720p at 24FPS with fluid motion and powerful prompt understanding
Wan 2.2's 5B distill model produces up to 5 seconds of video 720p at 24FPS with fluid motion and powerful prompt understanding
Wan 2.2's 5B FastVideo model produces up to 5 seconds of video 720p at 24FPS with fluid motion and powerful prompt understanding
Wan 2.2's 14B model edit high-resolution, photorealistic images with powerful prompt understanding and fine-grained visual detail
fal-ai/wan/v2.2-A14B/image-to-video
Wan-2.2 image-to-video is a video model that generates high-quality videos with high visual quality and motion diversity from text prompts and images. This endpoint supports LoRAs made for Wan 2.2
Wan-2.2 Turbo image-to-video is a video model that generates high-quality videos with high visual quality and motion diversity from text prompts.
Wan 2.2's 14B model generates high-resolution, photorealistic images with powerful prompt understanding and fine-grained visual detail
Wan 2.2's 14B model with LoRA support generates high-fidelity images with enhanced prompt alignment, style adaptability.
Wan-2.2 text-to-video is a video model that generates high-quality videos with high visual quality and motion diversity from text prompts.
Wan-2.2 text-to-video is a video model that generates high-quality videos with high visual quality and motion diversity from text prompts. This endpoint supports LoRAs made for Wan 2.2.
Wan-2.2 turbo text-to-video is a video model that generates high-quality videos with high visual quality and motion diversity from text prompts.
Wan-2.2 video-to-video is a video model that generates high-quality videos with high visual quality and motion diversity from text prompts and source videos.
Wan 2.6 image-to-image model.
Wan 2.6 image-to-video model.
Wan 2.6 image-to-video flash model.
Wan 2.6 reference-to-video flash model.
Wan 2.6 text-to-image model.
Wan 2.6 text-to-video model.
Transform and edit existing images with text-guided instructions using the WAN 2.7 model for creative image manipulation.
Wan 2.7 is the latest generation AI video model, delivering enhanced motion smoothness, superior scene fidelity, and greater visual coherence.
Wan 2.7 is the latest generation AI video model, delivering enhanced motion smoothness, superior scene fidelity, and greater visual coherence.
Edit and transform images using text instructions with the WAN 2.7 Pro model for precise, professional-grade image modifications.
Generate premium-quality images from text prompts using the enhanced WAN 2.7 Pro model with superior detail and composition.
Wan 2.7 is the latest generation AI video model, delivering enhanced motion smoothness, superior scene fidelity, and greater visual coherence.
Generate high-quality images from text prompts using the WAN 2.7 model with advanced prompt understanding and detailed output.
Wan 2.7 is the latest generation AI video model, delivering enhanced motion smoothness, superior scene fidelity, and greater visual coherence.
VACE is a video generation model that uses a source image, mask, and video to create prompted videos with controllable sources.
VACE is a video generation model that uses a source image, mask, and video to create prompted videos with controllable sources.
VACE is a video generation model that uses a source image, mask, and video to create prompted videos with controllable sources.
VACE is a video generation model that uses a source image, mask, and video to create prompted videos with controllable sources.
VACE is a video generation model that uses a source image, mask, and video to create prompted videos with controllable sources.
VACE is a video generation model that uses a source image, mask, and video to create prompted videos with controllable sources.
Reframe entire videos scene-by-scene using Wan VACE 2.1
Edit videos using plain language and Wan VACE
[Experimental] Whisper v3 Large -- but optimized by our inference wizards. Same WER, double the performance!
FFMPEG Utility for Audio Compression
Add automatic subtitles to videos
FFMPEG Utility for Blending Videos
FFMPEG Untility for Extracting nth Frame
FFMPEG Utility for Impulse Response
ffmpeg utility to interleave videos
Choose the Nth image from an image URL list for workflows.
FFMPEG Utility to Reverse Videos
FFMPEG Utilities to Scale Videos
FFMPEG Utility for Trim Video
Predict whether an image is NSFW or SFW.
Generate highly aesthetic images with xAI's Grok Imagine Image generation model.
Edit images precisely with xAI's Grok Imagine model
Grok Imagine Pro is an advanced AI model from xAI that creates high-quality visuals from text prompts and allows you to edit or analyze existing images.
Grok Imagine Pro is an advanced AI model from xAI that creates high-quality visuals from text prompts and allows you to edit or analyze existing images.
Edit images with xAi's Grok Imagine 2.0 model.
Generate images from text using xAi's Grok Imagine 2.0 model.
Edit videos using xAI's Grok Imagine
Extend videos with xAI's Grok Imagine video model
Generate videos from images with audio using xAI's Grok Imagine Video model.
Generate videos using multiple reference images with xAI's Grok Imagine video model
Generate videos with audio from text using Grok Imagine Video.
Generate videos from images with audio using xAI's Grok Imagine 1.5 Video model.
Generate videos from images using xAI's Grok Imagine Video 1.5 Lite model.
Generate videos from text prompts using xAI's Grok Imagine Video 1.5 Lite model.
Generate videos from images and audio references using xAI's Grok Imagine 1.5 Video model.
Generate videos from prompts with audio using xAI's Grok Imagine 1.5 Video model.
Build real-time voice applications powered by Grok. Send recorded audio and get a voice and text response.
Build real-time voice applications powered by Grok. Stream audio and text bidirectionally via WebSocket for voice assistants, phone agents, and interactive voice systems.
Generate speech with expressive and realistic voices from xAI
Z-Image is the foundation model of the Z- Image family, engineered for good quality, robust generative diversity, broad stylistic coverage, and precise prompt adherence.
LoRA endpoint for Z-Image, the foundation model of the Z- Image family.
Train LoRAs on Z-Image Turbo, a super fast text-to-image model of 6B parameters developed by Tongyi-MAI.
Z-Image Turbo is a super fast text-to-image model of 6B parameters developed by Tongyi-MAI.
Generate images from text and edge, depth or pose images using Z-Image Turbo, Tongyi-MAI's super-fast 6B model.
Generate images from text and edge, depth or pose images using custom LoRA and Z-Image Turbo, Tongyi-MAI's super-fast 6B model.
Generate images from text and images using Z-Image Turbo, Tongyi-MAI's super-fast 6B model.
Generate images from text and images using custom LoRA and Z-Image Turbo, Tongyi-MAI's super-fast 6B model.
Generate images from text, an image and a mask using Z-Image Turbo, Tongyi-MAI's super-fast 6B model.
Generate images from text, an image, a mask and custom LoRA using Z-Image Turbo, Tongyi-MAI's super-fast 6B model.
Text-to-Image endpoint with LoRA support for Z-Image Turbo, a super fast text-to-image model of 6B parameters developed by Tongyi-MAI.
Generate seamlessly tiling photorealistic images from text using Z-Image Turbo
Generate seamlessly tiling photorealistic images from text using Z-Image Turbo and custom LoRA
Zonos2 is a text-to-speech model that clones a voice from a short sample and speaks naturally across many languages.
Aion-2.0 is a variant of DeepSeek V3.2 optimized for immersive roleplaying and storytelling. It is particularly strong at introducing tension, crises, and conflict into stories, making narratives feel more engaging....
Aion-3.0 is a multi-model roleplaying and storytelling system from AionLabs, built on the GLM family of models. It uses a collaborative generation process in which multiple specialized models each contribute...
Aion-3.0 Mini is a multi-model roleplaying and storytelling system from AionLabs, built on the DeepSeek family of models. It uses a collaborative generation process in which multiple specialized models each...
Aion 3.5 is a multi-model roleplaying and storytelling system from AionLabs, built on the GLM family of models. It uses a collaborative generation process in which multiple specialized models each...
Aion 3.5 Mini is a multi-model roleplaying and storytelling system from AionLabs, built on the GLM family of models. It is the smaller, lower-cost sibling of Aion 3.5 and uses...
Aion-RP-Llama-3.1-8B ranks the highest in the character evaluation portion of the RPBench-Auto benchmark, a roleplaying-specific variant of Arena-Hard-Auto, where LLMs evaluate each other’s responses. It is a fine-tuned base model...
Nova 2 Lite is a fast, cost-effective reasoning model for everyday workloads that can process text, images, and videos to generate text. Nova 2 Lite demonstrates standout capabilities in processing...
Amazon Nova Lite 1.0 is a very low-cost multimodal model from Amazon that focused on fast processing of image, video, and text inputs to generate text output. Amazon Nova Lite...
Amazon Nova Micro 1.0 is a text-only model that delivers the lowest latency responses in the Amazon Nova family of models at a very low cost. With a context length...
Amazon Nova Premier is the most capable of Amazon’s multimodal models for complex reasoning tasks and for use as the best teacher for distilling custom models.
Amazon Nova Pro 1.0 is a capable multimodal model from Amazon focused on providing a combination of accuracy, speed, and cost for a wide range of tasks. As of December...
This is a series of models designed to replicate the prose quality of the Claude 3 models, specifically Sonnet(https://openrouter.ai/anthropic/claude-3.5-sonnet) and Opus(https://openrouter.ai/anthropic/claude-3-opus). The model is fine-tuned on top of [Qwen2.5 72B](https://openrouter.ai/qwen/qwen-2.5-72b-instruct).
Claude Fable 5 is a Mythos-class model from Anthropic, built for autonomous knowledge work and coding. It supports text, image, and file inputs with text output, with reasoning support and...
Claude Fable 5.1 improves on Claude Fable 5 across the board, with the biggest gains in agentic coding, long-running agentic workflows, and knowledge work: long code refactors, front-end and visual...
Claude Fable 5.1 improves on Claude Fable 5 across the board, with the biggest gains in agentic coding, long-running agentic workflows, and knowledge work: long code refactors, front-end and visual...
Claude Fable 5.2 is Anthropic's most capable model for the most demanding technical work, with gains over Fable 5.1 in coding, reasoning, and vision. In prerelease testing it completed otherwise-infeasible...
Claude Fable 5.2 is Anthropic's most capable model for the most demanding technical work, with gains over Fable 5.1 in coding, reasoning, and vision. In prerelease testing it completed otherwise-infeasible...
Claude Fable 5 is a Mythos-class model from Anthropic, built for autonomous knowledge work and coding. It supports text, image, and file inputs with text output, with reasoning support and...
This model always redirects to the latest model in the Claude Fable family.
Claude Haiku 4.5 is Anthropic’s fastest and most efficient model, delivering near-frontier intelligence at a fraction of the cost and latency of larger Claude models. Matching Claude Sonnet 4’s performance...
Claude Haiku 4.5 is Anthropic’s fastest and most efficient model, delivering near-frontier intelligence at a fraction of the cost and latency of larger Claude models. Matching Claude Sonnet 4’s performance...
Claude Haiku 5.5 is Anthropic's small, fast model for high-volume, cost-sensitive work such as summarization, subagents, and browser use. It succeeds Claude Haiku 4.5 with stronger coding, computer use, and...
Claude Haiku 5.5 is Anthropic's small, fast model for high-volume, cost-sensitive work such as summarization, subagents, and browser use. It succeeds Claude Haiku 4.5 with stronger coding, computer use, and...
This model always redirects to the latest model in the Claude Haiku family.
Claude Opus 4.1 is an updated version of Anthropic’s flagship model, offering improved performance in coding, reasoning, and agentic tasks. It achieves 74.5% on SWE-bench Verified and shows notable gains...
Claude Opus 4.1 is an updated version of Anthropic’s flagship model, offering improved performance in coding, reasoning, and agentic tasks. It achieves 74.5% on SWE-bench Verified and shows notable gains...
Claude Opus 4.5 is Anthropic’s frontier reasoning model optimized for complex software engineering, agentic workflows, and long-horizon computer use. It offers strong multimodal capabilities, competitive performance across real-world coding and...
Claude Opus 4.5 is Anthropic’s frontier reasoning model optimized for complex software engineering, agentic workflows, and long-horizon computer use. It offers strong multimodal capabilities, competitive performance across real-world coding and...
Opus 4.6 is Anthropic’s strongest model for coding and long-running professional tasks. It is built for agents that operate across entire workflows rather than single prompts, making it especially effective...
Opus 4.6 is Anthropic’s strongest model for coding and long-running professional tasks. It is built for agents that operate across entire workflows rather than single prompts, making it especially effective...
Opus 4.7 is the next generation of Anthropic's Opus family, built for long-running, asynchronous agents. Building on the coding and agentic strengths of Opus 4.6, it delivers stronger performance on...
Opus 4.7 is the next generation of Anthropic's Opus family, built for long-running, asynchronous agents. Building on the coding and agentic strengths of Opus 4.6, it delivers stronger performance on...
Claude Opus 4.8 is Anthropic's most capable generally available model in the Opus family. It supports text, image, and file inputs with text output, with reasoning support and a 1M-token...
Claude Opus 4.8 is Anthropic's most capable generally available model in the Opus family. It supports text, image, and file inputs with text output, with reasoning support and a 1M-token...
Claude Opus 5 is Anthropic’s flagship model for demanding reasoning, coding, and long-horizon agentic work. It is particularly strong at end-to-end software tasks, code review and bug finding, visual analysis...
Claude Opus 5.5 is Anthropic's flagship model for demanding reasoning, coding, and long-horizon agentic work, succeeding Claude Opus 5. It is particularly strong at multi-step changes in large codebases, code...
Claude Opus 5.5 is Anthropic's flagship model for demanding reasoning, coding, and long-horizon agentic work, succeeding Claude Opus 5. It is particularly strong at multi-step changes in large codebases, code...
Claude Opus 5 is Anthropic’s flagship model for demanding reasoning, coding, and long-horizon agentic work. It is particularly strong at end-to-end software tasks, code review and bug finding, visual analysis...
This model always redirects to the latest model in the Claude Opus family.
Claude Sonnet 4.5 is Anthropic’s most advanced Sonnet model to date, optimized for real-world agents and coding workflows. It delivers state-of-the-art performance on coding benchmarks such as SWE-bench Verified, with...
Claude Sonnet 4.5 is Anthropic’s most advanced Sonnet model to date, optimized for real-world agents and coding workflows. It delivers state-of-the-art performance on coding benchmarks such as SWE-bench Verified, with...
Sonnet 4.6 is Anthropic's most capable Sonnet-class model yet, with frontier performance across coding, agents, and professional work. It excels at iterative development, complex codebase navigation, end-to-end project management with...
Sonnet 4.6 is Anthropic's most capable Sonnet-class model yet, with frontier performance across coding, agents, and professional work. It excels at iterative development, complex codebase navigation, end-to-end project management with...
Sonnet 5 is Anthropic's most capable Sonnet-class model, with frontier performance across coding, agents, and professional work. It supports adaptive thinking with selectable reasoning effort levels (low, medium, high, max,...
Claude Sonnet 5.5 is Anthropic's Sonnet-class model for well-scoped everyday work, succeeding Claude Sonnet 5 as a direct upgrade. It is especially strong at building features, fixing bugs, and producing...
Claude Sonnet 5.5 is Anthropic's Sonnet-class model for well-scoped everyday work, succeeding Claude Sonnet 5 as a direct upgrade. It is especially strong at building features, fixing bugs, and producing...
Sonnet 5 is Anthropic's most capable Sonnet-class model, with frontier performance across coding, agents, and professional work. It supports adaptive thinking with selectable reasoning effort levels (low, medium, high, max,...
This model always redirects to the latest model in the Claude Sonnet family.
ERNIE 5.5 Flash Preview is the latest foundation model in the ERNIE family, with comprehensive improvements in coding, agentic capabilities, knowledge work, and deep research. It has 280 billion total...
ERNIE-Light-DT-1.0 is based on open-source GLM models and includes research findings from Baidu Search's Lightning large model. It features Baidu Sparse Attention (BSA), a unique sparse attention mechanism developed by...
Seed-2.0-Lite is a versatile, cost‑efficient enterprise workhorse that delivers strong multimodal and agent capabilities while offering noticeably lower latency, making it a practical default choice for most production workloads across...
Seed-2.0-mini targets latency-sensitive, high-concurrency, and cost-sensitive scenarios, emphasizing fast response and flexible inference deployment. It delivers performance comparable to ByteDance-Seed-1.6, supports 256k context, four reasoning effort modes (minimal/low/medium/high), multimodal understanding,...
Seed 2.1 Turbo is a multimodal model from ByteDance Seed for coding and long-horizon agent workflows. It is suited for end-to-end software delivery, multi-step task execution, and understanding visual and...
UI-TARS-1.5 is a multimodal vision-language agent optimized for GUI-based environments, including desktop interfaces, web browsers, mobile systems, and games. Built by ByteDance, it builds upon the UI-TARS framework with reinforcement...
Venice Uncensored Dolphin Mistral 24B Venice Edition is a fine-tuned variant of Mistral-Small-24B-Instruct-2501, developed by dphn.ai in collaboration with Venice.ai. This model is designed as an “uncensored” instruct-tuned LLM, preserving...
Command A is an open-weights 111B parameter model with a 256k context window focused on delivering great performance across agentic, multilingual, and coding use cases. Compared to other leading proprietary...
Command A+ is Cohere's flagship model for enterprise agentic workflows. It accepts text and image inputs with a 192K context window, supports native tool calling with strict tool schemas, structured...
command-r-08-2024 is an update of the [Command R](/models/cohere/command-r) with improved performance for multilingual retrieval-augmented generation (RAG) and tool use. More broadly, it is better at math, code and reasoning and...
command-r-plus-08-2024 is an update of the [Command R+](/models/cohere/command-r-plus) with roughly 50% higher throughput and 25% lower latencies as compared to the previous Command R+ version, while keeping the hardware footprint...
Command R7B (12-2024) is a small, fast update of the Command R+ model, delivered in December 2024. It excels at RAG, tool use, agents, and similar tasks requiring complex reasoning...
DeepSeek-V3 is the latest model from the DeepSeek team, building upon the instruction following and coding abilities of the previous versions. Pre-trained on nearly 15 trillion tokens, the reported evaluations...
DeepSeek V3, a 685B-parameter, mixture-of-experts model, is the latest iteration of the flagship chat model family from the DeepSeek team. It succeeds the [DeepSeek V3](/deepseek/deepseek-chat-v3) model and performs really well...
DeepSeek-V3.1 is a large hybrid reasoning model (671B parameters, 37B active) that supports both thinking and non-thinking modes via prompt templates. It extends the DeepSeek-V3 base with a two-phase long-context...
This model always redirects to the latest model in the DeepSeek Flash family.
This model always redirects to the latest model in the DeepSeek Pro family.
May 28th update to the [original DeepSeek R1](/deepseek/deepseek-r1) Performance on par with [OpenAI o1](/openai/o1), but open-sourced and with fully open reasoning tokens. It's 671B parameters in size, with 37B active...
DeepSeek-V3.1 Terminus is an update to [DeepSeek V3.1](/deepseek/deepseek-chat-v3.1) that maintains the model's original capabilities while addressing issues reported by users, including language consistency and agent capabilities, further optimizing the model's...
DeepSeek-V3.2 is a large language model designed to harmonize high computational efficiency with strong reasoning and agentic tool-use performance. It introduces DeepSeek Sparse Attention (DSA), a fine-grained sparse attention mechanism...
DeepSeek-V3.2-Exp is an experimental large language model released by DeepSeek as an intermediate step between V3.1 and future architectures. It introduces DeepSeek Sparse Attention (DSA), a fine-grained sparse attention mechanism...
DeepSeek V4.1 Flash is a sparse mixture-of-experts model from DeepSeek, and the first built on the company's Causal Encoder-Decoder (CED) architecture. It activates 8B parameters on input and 16B on...
DeepSeek V4.1 Flash 1017 is the October 17, 2026 snapshot of DeepSeek V4.1 Flash, served by DeepSeek under the `deepseek-flash` API model name. V4.1 Flash is a sparse mixture-of-experts model...
DeepSeek V4.1 Flash 1017 is the October 17, 2026 snapshot of DeepSeek V4.1 Flash, served by DeepSeek under the `deepseek-flash` API model name. V4.1 Flash is a sparse mixture-of-experts model...
DeepSeek V4.1 Flash is a sparse mixture-of-experts model from DeepSeek, and the first built on the company's Causal Encoder-Decoder (CED) architecture. It activates 8B parameters on input and 16B on...
DeepSeek V4 Flash is an efficiency-optimized Mixture-of-Experts model from DeepSeek with 284B total parameters and 13B activated parameters, supporting a 1M-token context window. It is designed for fast inference and...
DeepSeek V4 Flash 0731 is a sparse mixture-of-experts model from DeepSeek, with 13B active parameters out of 284B total. This re-post-trained revision is suited for coding, reasoning, and agent workflows....
DeepSeek V4 Flash 0731 is a sparse mixture-of-experts model from DeepSeek, with 13B active parameters out of 284B total. This re-post-trained revision is suited for coding, reasoning, and agent workflows....
This model always redirects to the latest model in the DeepSeek V4 Flash family.
DeepSeek V4 Flash Vision Exp is an experimental vision-enabled version of [DeepSeek V4 Flash 0731](https://openrouter.ai/deepseek/deepseek-v4-flash-0731) from DeepSeek, adding image understanding while matching the base model on text capabilities including agents,...
DeepSeek V4 Pro is a large-scale Mixture-of-Experts model from DeepSeek with 1.6T total parameters and 49B activated parameters, supporting a 1M-token context window. It is designed for advanced reasoning, coding,...
DeepSeek V4 Pro 0813 is a large-scale mixture-of-experts model from DeepSeek. This is the GA release of DeepSeek V4 Pro.
Glimpse is a multimodal visual reasoning model from Elorian, built for visual understanding and complex visual reasoning in the physical world. It reasons directly over images and video rather than...
Ember-1 is a specialized reasoning model from Fireworks Research, built on [Kimi K3](https://openrouter.ai/moonshotai/kimi-k3). It is designed to make every token go further: it produces shorter reasoning traces, using roughly 40%...
Gemini 2.5 Flash is Google's state-of-the-art workhorse model, specifically designed for advanced reasoning, coding, mathematics, and scientific tasks. It includes built-in "thinking" capabilities, enabling it to provide responses with greater...
Gemini 2.5 Flash Image, a.k.a. "Nano Banana," is now generally available. It is a state of the art image generation model with contextual understanding. It is capable of image generation,...
Gemini 2.5 Flash-Lite is a lightweight reasoning model in the Gemini 2.5 family, optimized for ultra-low latency and cost efficiency. It offers improved throughput, faster token generation, and better performance...
Gemini 2.5 Pro is Google’s state-of-the-art AI model designed for advanced reasoning, coding, mathematics, and scientific tasks. It employs “thinking” capabilities, enabling it to reason through responses with enhanced accuracy...
Gemini 2.5 Pro is Google’s state-of-the-art AI model designed for advanced reasoning, coding, mathematics, and scientific tasks. It employs “thinking” capabilities, enabling it to reason through responses with enhanced accuracy...
Gemini 3.1 Flash Image, a.k.a. "Nano Banana 2," is Google’s latest state of the art image generation and editing model, delivering Pro-level visual quality at Flash speed. It combines advanced...
Gemini 3.1 Flash Image Preview, a.k.a. "Nano Banana 2," is Google’s latest state of the art image generation and editing model, delivering Pro-level visual quality at Flash speed. It combines...
Gemini 3.1 Flash Lite is Google’s GA high-efficiency multimodal model optimized for low-latency, high-volume workloads. It supports text, image, video, audio, and PDF inputs, and is designed for lightweight agentic...
Gemini 3.1 Flash Lite is Google’s GA high-efficiency multimodal model optimized for low-latency, high-volume workloads. It supports text, image, video, audio, and PDF inputs, and is designed for lightweight agentic...
Nano Banana 2 Lite (Gemini 3.1 Flash Lite Image) is Google's fastest, most cost-efficient Gemini image model, built for high-velocity developer pipelines and rapid-fire visual exploration. It delivers text-to-image generation...
Gemini 3.1 Flash Lite Preview is Google's high-efficiency model optimized for high-volume use cases. It outperforms Gemini 2.5 Flash Lite on overall quality and approaches Gemini 2.5 Flash performance across...
Gemini 3.1 Pro Preview is Google’s frontier reasoning model, delivering enhanced software engineering performance, improved agentic reliability, and more efficient token usage across complex workflows. Building on the multimodal foundation...
Gemini 3.1 Pro Preview is Google’s frontier reasoning model, delivering enhanced software engineering performance, improved agentic reliability, and more efficient token usage across complex workflows. Building on the multimodal foundation...
Gemini 3.1 Pro Preview Custom Tools is a variant of Gemini 3.1 Pro that improves tool selection behavior by preventing overuse of a general bash tool when more efficient third-party...
Gemini 3.5 Flash is Google's high-efficiency multimodal model, bringing near-Pro level coding and reasoning at Flash-tier cost and speed. It is highly optimized for coding proficiency and parallel agentic execution...
Gemini 3.5 Flash is Google's high-efficiency multimodal model, bringing near-Pro level coding and reasoning at Flash-tier cost and speed. It is highly optimized for coding proficiency and parallel agentic execution...
Gemini 3.5 Flash Lite is a high-efficiency model from Google with upgraded agentic capabilities. It is suited for subagents that execute focused tasks within complex, multi-agent workflows.
Gemini 3.5 Flash Lite is a high-efficiency model from Google with upgraded agentic capabilities. It is suited for subagents that execute focused tasks within complex, multi-agent workflows.
Gemini 3.6 Flash is a high-efficiency model from Google for coding, agentic workflows, and web and app development. It is designed to produce polished outputs with fewer unnecessary edits and...
Gemini 3.6 Flash is a high-efficiency model from Google for coding, agentic workflows, and web and app development. It is designed to produce polished outputs with fewer unnecessary edits and...
Gemini 3.7 Flash is a multimodal model from Google for fast agentic workflows, coding, and complex multi-step reasoning. It is designed for tasks that require responsive performance and reliable multi-step...
Gemini 3.7 Flash is a multimodal model from Google for fast agentic workflows, coding, and complex multi-step reasoning. It is designed for tasks that require responsive performance and reliable multi-step...
Gemini 3.8 Flash is Google's most intelligent Flash model with significant gains from 3.7 Flash across software engineering, agentic tasks, and multi-step reasoning.
Gemini 3.8 Flash is Google's most intelligent Flash model with significant gains from 3.7 Flash across software engineering, agentic tasks, and multi-step reasoning.
Gemini 3 Flash Preview is a high speed, high value thinking model designed for agentic workflows, multi turn chat, and coding assistance. It delivers near Pro level reasoning and tool...
Gemini 3 Flash Preview is a high speed, high value thinking model designed for agentic workflows, multi turn chat, and coding assistance. It delivers near Pro level reasoning and tool...
Nano Banana Pro is Google’s most advanced image-generation and editing model, built on Gemini 3 Pro. It extends the original Nano Banana with significantly improved multimodal reasoning, real-world grounding, and...
Nano Banana Pro is Google’s most advanced image-generation and editing model, built on Gemini 3 Pro. It extends the original Nano Banana with significantly improved multimodal reasoning, real-world grounding, and...
Gemini 4 Argon is Google's flagship reasoning model, built for complex software engineering, deep mathematical and scientific problem solving, and autonomous multi-step agents. Google trained it to keep context across...
Gemini 4 Argon is Google's flagship reasoning model, built for complex software engineering, deep mathematical and scientific problem solving, and autonomous multi-step agents. Google trained it to keep context across...
Gemini 4 Carbon is a frontier Gemini 4 model from Google for coding, agentic and complex reasoning work. It takes text, image, audio, video and file input, supports long decode...
Gemini 4 Carbon is a frontier Gemini 4 model from Google for coding, agentic and complex reasoning work. It takes text, image, audio, video and file input, supports long decode...
Gemini 4 Hydrogen is the high-efficiency workhorse model of the Gemini 4 family from Google, built for high-throughput agentic workflows, coding and multi-model routing at lower cost and latency than...
Gemini 4 Hydrogen is the high-efficiency workhorse model of the Gemini 4 family from Google, built for high-throughput agentic workflows, coding and multi-model routing at lower cost and latency than...
Gemini 4 Neon is a fast, mid-priced model in the Gemini 4 family from Google, built for coding and agentic work. It is suited for repository and terminal coding, web...
Gemini 4 Neon is a fast, mid-priced model in the Gemini 4 family from Google, built for coding and agentic work. It is suited for repository and terminal coding, web...
This model always redirects to the latest model in the Gemini Flash family.
Nano Banana 2.1 (Gemini Nano Banana 2.1) is Google's image generation and editing model on the Flash tier, succeeding Nano Banana 2 and Nano Banana Pro. It improves product recontextualization,...
Gemini Omni 1.1 Flash is Google's native multimodal video generation and editing model, the generally available successor to Gemini Omni Flash Preview. It is suited for text-to-video, image-to-video, first and...
Gemini Omni 1.2 Flash is Google's native multimodal video generation and editing model, the successor to Gemini Omni 1.1 Flash. It generates clips up to 15 seconds long with native...
This model always redirects to the latest model in the Gemini Pro family.
Gemma 2 27B by Google is an open model built from the same research and technology used to create the [Gemini models](/models?q=gemini). Gemma models are well-suited for a variety of...
Gemma 3 introduces multimodality, supporting vision-language input and text outputs. It handles context windows up to 128k tokens, understands over 140 languages, and offers improved math, reasoning, and chat capabilities,...
Gemma 3 introduces multimodality, supporting vision-language input and text outputs. It handles context windows up to 128k tokens, understands over 140 languages, and offers improved math, reasoning, and chat capabilities,...
Gemma 3 introduces multimodality, supporting vision-language input and text outputs. It handles context windows up to 128k tokens, understands over 140 languages, and offers improved math, reasoning, and chat capabilities,...
Gemma 4 26B A4B IT is an instruction-tuned Mixture-of-Experts (MoE) model from Google DeepMind. Despite 25.2B total parameters, only 3.8B activate per token during inference — delivering near-31B quality at...
Gemma 4 26B A4B IT is an instruction-tuned Mixture-of-Experts (MoE) model from Google DeepMind. Despite 25.2B total parameters, only 3.8B activate per token during inference — delivering near-31B quality at...
Gemma 4 31B Instruct is Google DeepMind's 30.7B dense multimodal model supporting text and image input with text output. Features a 256K token context window, configurable thinking/reasoning mode, native function...
Gemma 4 31B Instruct is Google DeepMind's 30.7B dense multimodal model supporting text and image input with text output. Features a 256K token context window, configurable thinking/reasoning mode, native function...
30 second duration clips are priced at $0.04 per clip. Lyria 3 is Google's family of music generation models, available through the Gemini API. With Lyria 3, you can generate...
Full-length songs are priced at $0.08 per song. Lyria 3 is Google's family of music generation models, available through the Gemini API. With Lyria 3, you can generate high-quality, 48kHz...
One of the highest performing and most popular fine-tunes of Llama 2 13B, with rich descriptions and roleplay. #merge
Granite 4.2 8B is a dense reasoning model from IBM. It is suited for mathematics, code generation, multilingual dialogue, and agentic workflows that need multi-step reasoning. It supports full, low-effort,...
*Ling-3.0-flash* is a *124B-parameter Mixture-of-Experts (MoE) model*, with approximately *5.1B parameters activated per token*. The model is designed with *token efficiency and production-scale agentic inference* as key priorities, enabling developers...
Ling 3.0 Flash VL builds on Ling 3.0 Flash (124B total / 5.5B active MoE from InclusionAI), further strengthening its language capabilities while adding native visual perception and advanced visual...
Ling 3.0 Tiny is a mixture-of-experts model from InclusionAI, with 1.3B active parameters out of 7.9B total. It is designed for responsive agents, instruction following, and multi-turn conversations, with switchable...
Ling 3.1 Flash is a hybrid reasoning mixture-of-experts model from inclusionAI, with 25B active parameters out of 560B total.
Ling 3.2 Flash is a hybrid reasoning model from inclusionAI that accepts text and image input, with a 1M-token context window and up to 200 images per request. Thinking can...
Schematron V2 Small is a 3B-parameter HTML-to-JSON extraction model from Inference.net. It prioritizes extraction quality for complex schemas and long pages. Extraction instructions must be supplied through a JSON schema...
Schematron V2 Turbo is a 3B-parameter HTML-to-JSON extraction model from Inference.net. It prioritizes throughput for high-volume extraction workloads. Extraction instructions must be supplied through a JSON schema in response_format rather...
Mi:dm K 3.0 Pro is a Mixture-of-Experts reasoning model from KT, with 120B total and 12B active parameters. It uses a hybrid architecture that mixes Mamba, MoE and attention layers.
LFM2.5-24B-A2B is a mixture-of-experts reasoning model from Liquid AI with 24B total and 2.45B active parameters. It always reasons before answering and is built for agents that call tools. Liquid...
A compact 1.7B-parameter text model with a 256K-token context, post-trained for agents, tool calling, and structured outputs. It reasons before it answers. Prompts and outputs may be retained and used...
A compact 1.7B-parameter text model with a 256K-token context, post-trained for agents, tool calling, and structured outputs. It reasons before it answers. Prompts and outputs may be retained and used...
An attempt to recreate Claude-style verbosity, but don't expect the same level of coherence or memory. Meant for use in roleplay/narrative situations.
Meta's latest class of model (Llama 3.1) launched with a variety of sizes & flavors. This 70B instruct-tuned version is optimized for high quality dialogue usecases. It has demonstrated strong...
Meta's latest class of model (Llama 3.1) launched with a variety of sizes & flavors. This 8B instruct-tuned version is fast and efficient. It has demonstrated strong performance compared to...
Llama 3.2 3B is a 3-billion-parameter multilingual large language model, optimized for advanced natural language processing tasks like dialogue generation, reasoning, and summarization. Designed with the latest transformer architecture, it...
The Meta Llama 3.3 multilingual large language model (LLM) is a pretrained and instruction tuned generative model in 70B (text in/text out). The Llama 3.3 instruction tuned text only model...
Llama 4 Maverick 17B Instruct (128E) is a high-capacity multimodal language model from Meta, built on a mixture-of-experts (MoE) architecture with 128 experts and 17 billion active parameters per forward...
Llama 4 Scout 17B Instruct (16E) is a mixture-of-experts (MoE) language model developed by Meta, activating 17 billion parameters out of a total of 109B. It supports native multimodal input...
Llama Guard 4 is a Llama 4 Scout-derived multimodal pretrained model, fine-tuned for content safety classification. Similar to previous versions, it can be used to classify content in both LLM...
Muse Ember is the smallest and most efficient model from Meta that’s designed for complex reasoning execution and the work agents repeat all day: tool calls, coding agent subtasks, routing,...
Muse Glimmer 30B is a dense, open-weight multimodal model from Meta Superintelligence Labs, distilled from Muse Spark and optimized for autonomous agents on consumer hardware. It is suited for long-horizon...
Muse Glimmer 30B is a dense, open-weight multimodal model from Meta Superintelligence Labs, distilled from Muse Spark and optimized for autonomous agents on consumer hardware. It is suited for long-horizon...
Muse Spark 1.1 is a multimodal reasoning model from Meta, built for agentic tasks. It accepts text, images, video, and PDF documents and returns text, with a 1M-token context window....
Muse Spark 1.2 is a reasoning model from Meta, designed for complex agentic tasks. It accepts text, images, video, and PDF documents, returns text, and offers a 1M-token context window....
Muse Spark 1.2 contributor tier is a reasoning model from Meta designed for developers who want to start building at an even lower cost. It’s meaningfully cheaper than Muse Spark...
Muse Spark 1.3 is a multimodal reasoning model from Meta for long-running agentic, multi-agent, and coding workflows. It is designed to keep track of information across extended tasks, work through...
Muse Spark 1.3 Contributor is the cost-efficient contributor tier of Meta’s multimodal reasoning model for experimentation, learning, and early-stage agentic, multi-agent, and coding workflows. It is designed to track information...
Muse Spark Open is a coding-optimized open-weight model. Trained specifically for agentic coding workflows, it delivers higher first-attempt accuracy, reliable tool calling, and deep codebase understanding across the repos developers...
Muse Wonder is a frontier reasoning model from Meta, designed for complex personal and professional agentic tasks. It accepts text, images, video, audio, and PDF documents and returns text, with...
[Microsoft Research](/microsoft) Phi-4 is designed to perform well in complex reasoning tasks and can operate efficiently in situations with limited memory or where quick responses are needed. At 14 billion...
MiniMax-M2.5 is a SOTA large language model designed for real-world productivity. Trained in a diverse range of complex real-world digital working environments, M2.5 builds upon the coding expertise of M2.1...
MiniMax-M2.7 is a next-generation large language model designed for autonomous, real-world productivity and continuous improvement. Built to actively participate in its own evolution, M2.7 integrates advanced agentic capabilities through multi-agent...
MiniMax-M3 is a multimodal foundation model from MiniMax. It supports text, image, and video inputs with text output, a 1M-token context window, and is suited for long-horizon agentic work, coding,...
MiniMax-M3.1-Flash is a multimodal model from MiniMax. Compared with M3, it delivers significant improvements across multiple benchmarks, particularly in complex agentic tasks, coding, and cybersecurity. It is built for complex,...
MiniMax-M3 is a multimodal foundation model from MiniMax. It supports text, image, and video inputs with text output, a 1M-token context window, and is suited for long-horizon agentic work, coding,...
Mistral's cutting-edge language model for coding released end of July 2025. Codestral specializes in low-latency, high-frequency tasks such as fill-in-the-middle (FIM), code correction and test generation. [Blog Post](https://mistral.ai/news/codestral-25-08)
Devstral 2 is a state-of-the-art open-source model by Mistral AI specializing in agentic coding. It is a 123B-parameter dense transformer model supporting a 256K context window. Devstral 2 supports exploring...
The largest model in the Ministral 3 family, Ministral 3 14B offers frontier capabilities and performance comparable to its larger Mistral Small 3.2 24B counterpart. A powerful and efficient language...
The smallest model in the Ministral 3 family, Ministral 3 3B is a powerful, efficient tiny language model with vision capabilities.
A balanced model in the Ministral 3 family, Ministral 3 8B is a powerful, efficient tiny language model with vision capabilities.
This is Mistral AI's flagship model, Mistral Large 2 (version `mistral-large-2407`). It's a proprietary weights-available model and excels at reasoning, code, JSON, chat, and more. Read the launch announcement [here](https://mistral.ai/news/mistral-large-2407/)....
This is Mistral AI's flagship model, Mistral Large 2 (version mistral-large-2407). It's a proprietary weights-available model and excels at reasoning, code, JSON, chat, and more. Read the launch announcement [here](https://mistral.ai/news/mistral-large-2407/)....
Mistral Large 3 2512 is Mistral’s most capable model to date, featuring a sparse mixture-of-experts architecture with 41B active parameters (675B total), and released under the Apache 2.0 license.
Mistral Large 4 is a frontier multimodal (text and image input) model from Mistral AI built for reasoning, coding, and agentic workloads. It offers a 1M-token context window with up...
Mistral Medium 3 is a high-performance enterprise-grade language model designed to deliver frontier-level capabilities at significantly reduced operational cost. It balances state-of-the-art reasoning and multimodal performance with 8× lower cost...
Mistral Medium 3.1 is an updated version of Mistral Medium 3, which is a high-performance enterprise-grade language model designed to deliver frontier-level capabilities at significantly reduced operational cost. It balances...
Mistral Medium 3.5 is a dense 128B instruction-following model from Mistral AI. It supports text and image inputs with text output, and is designed for agentic workflows, coding, and complex...
Mistral Medium 3.5 is a dense 128B instruction-following model from Mistral AI. It supports text and image inputs with text output, and is designed for agentic workflows, coding, and complex...
A 12B parameter model with a 128k token context length built by Mistral in collaboration with NVIDIA. The model is multilingual, supporting English, French, German, Spanish, Italian, Portuguese, Chinese, Japanese,...
Mistral Saba is a 24B-parameter language model specifically designed for the Middle East and South Asia, delivering accurate and contextually relevant responses while maintaining efficient performance. Trained on curated regional...
Mistral Small 3 is a 24B-parameter language model optimized for low-latency performance across common AI tasks. Released under the Apache 2.0 license, it features both pre-trained and instruction-tuned versions designed...
Mistral Small 4 is the next major release in the Mistral Small family, unifying the capabilities of several flagship Mistral models into a single system. It combines strong reasoning from...
Mistral Small 3.1 24B Instruct is an upgraded variant of Mistral Small 3 (2501), featuring 24 billion parameters with advanced multimodal capabilities. It provides state-of-the-art performance in text-based reasoning and...
Mistral-Small-3.2-24B-Instruct-2506 is an updated 24B parameter model from Mistral optimized for instruction following, repetition reduction, and improved function calling. Compared to the 3.1 release, version 3.2 significantly improves accuracy on...
Mistral's official instruct fine-tuned version of [Mixtral 8x22B](/models/mistralai/mixtral-8x22b). It uses 39B active parameters out of 141B, offering unparalleled cost efficiency for its size. Its strengths include: - strong math, coding,...
Pixtral Large 2502 is the February 2025 version of Mistral AI's large multimodal model, following [Pixtral Large 2411](https://openrouter.ai/mistralai/pixtral-large-2411). It accepts text and image input, including documents, charts, and natural images,...
Voxtral Small is an enhancement of Mistral Small 3, incorporating state-of-the-art audio input capabilities while retaining best-in-class text performance. It excels at speech transcription, translation and audio understanding. Input audio...
Moondream 3.1 is a vision-language model from Moondream, the successor to Moondream 3 Preview, with 9B total parameters and 2B active (mixture-of-experts). It handles visual question answering, captioning, OCR, and...
Moondream 3 Preview is a vision-language model from Moondream with 9B total parameters and 2B active (mixture-of-experts). It handles visual question answering, captioning, OCR, and visual reasoning over images, with...
Kimi K2 Instruct is a large-scale Mixture-of-Experts (MoE) language model developed by Moonshot AI, featuring 1 trillion total parameters with 32 billion active per forward pass. It is optimized for...
Kimi K2 0905 is the September update of [Kimi K2 0711](moonshotai/kimi-k2). It is a large-scale Mixture-of-Experts (MoE) language model developed by Moonshot AI, featuring 1 trillion total parameters with 32...
Kimi K2.5 is Moonshot AI's native multimodal model, delivering state-of-the-art visual coding capability and a self-directed agent swarm paradigm. Built on Kimi K2 with continued pretraining over approximately 15T mixed...
Kimi K2.6 is Moonshot AI's next-generation multimodal model, designed for long-horizon coding, coding-driven UI/UX generation, and multi-agent orchestration. It handles complex end-to-end coding tasks across Python, Rust, and Go, and...
MoonshotAI: Kimi K2.7 Code is a coding-focused model in Moonshot AI's Kimi K2 family, built to complete end-to-end programming tasks reliably over long contexts. It uses a native multimodal mixture-of-experts...
Kimi K2 Thinking is Moonshot AI’s most advanced open reasoning model to date, extending the K2 series into agentic, long-horizon reasoning. Built on the trillion-parameter Mixture-of-Experts (MoE) architecture introduced in...
Kimi K3 is a 2.8T parameter open-weight multimodal reasoning model from Moonshot AI. It is suited for complex coding, knowledge work, and long-horizon agentic workflows, and is particularly strong at...
Kimi K3.1 Pro is Moonshot AI's flagship open-weight multimodal reasoning model and a post-trained successor to Kimi K3, with a 1M-token context window. It is built for long-horizon coding, 3D...
Kimi K3.1 Pro is Moonshot AI's flagship open-weight multimodal reasoning model and a post-trained successor to Kimi K3, with a 1M-token context window. It is built for long-horizon coding, 3D...
Kimi K3 is a 2.8T parameter open-weight multimodal reasoning model from Moonshot AI. It is suited for complex coding, knowledge work, and long-horizon agentic workflows, and is particularly strong at...
This model always redirects to the latest model in the Kimi family.
Morph's fastest apply model for code edits. ~10,500 tokens/sec with 96% accuracy for rapid code transformations. The model requires the prompt to be in the following format: <instruction>{instruction}</instruction> <code>{initial_code}</code> <update>{edit_snippet}</update>...
Morph's fastest apply model for code edits. ~10,500 tokens/sec with 96% accuracy for rapid code transformations. The model requires the prompt to be in the following format: <instruction>{instruction}</instruction> <code>{initial_code}</code> <update>{edit_snippet}</update>...
Morph's high-accuracy apply model for complex code edits. ~4,500 tokens/sec with 98% accuracy for precise code transformations. The model requires the prompt to be in the following format: <instruction>{instruction}</instruction> <code>{initial_code}</code>...
Morph's high-accuracy apply model for complex code edits. ~4,500 tokens/sec with 98% accuracy for precise code transformations. The model requires the prompt to be in the following format: <instruction>{instruction}</instruction> <code>{initial_code}</code>...
Motif 3 is a Mixture-of-Experts language model from Motif Technologies, with 314B total and 13.2B active parameters. It is a general-purpose multilingual model for English and Korean, particularly strong at...
Nex-N2.5 is an agentic model built to turn goals into working, verified outcomes. Its core strength is agentic coding within a visual feedback loop: it can explore codebases, implement multi-file...
Nex-N2.5 is an agentic model built to turn goals into working, verified outcomes. Its core strength is agentic coding within a visual feedback loop: it can explore codebases, implement multi-file...
Nex-N2.6 is an agentic model from Nex AGI, built on [GLM-5.3-Flash](https://huggingface.co/zai-org/GLM-5.3-Flash-BF16) with about 321B parameters. It is suited for computer use, coding, and long-horizon tasks that involve tools. Compared with...
Hermes 4 is a large-scale reasoning model built on Meta-Llama-3.1-405B and released by Nous Research. It introduces a hybrid reasoning mode, where the model can choose to deliberate internally with...
NVIDIA Nemotron 3.5 Content Safety is a compact 4B-parameter multimodal guardrail model from NVIDIA, fine-tuned from Google Gemma-3-4B. It moderates both inputs to and responses from LLMs and VLMs, accepting...
NVIDIA Nemotron 3.5 Lightning is an open mixture-of-experts model from NVIDIA, with 3B active parameters out of 30B total. It is suited for high-throughput agentic workloads and specialized tasks that...
NVIDIA Nemotron 3.5 Lightning is an open mixture-of-experts model from NVIDIA, with 3B active parameters out of 30B total. It is suited for high-throughput agentic workloads and specialized tasks that...
NVIDIA Nemotron 3.5 Super VL is an open multimodal mixture-of-experts model from NVIDIA, with 12B active parameters out of 120B total. It takes text, images, documents, GUI screenshots, and video...
NVIDIA Nemotron 3.5 Super VL is an open multimodal mixture-of-experts model from NVIDIA, with 12B active parameters out of 120B total. It takes text, images, documents, GUI screenshots, and video...
NVIDIA Nemotron 3 Nano 30B A3B is a small language MoE model with highest compute efficiency and accuracy for developers to build specialized agentic AI systems. The model is fully...
NVIDIA Nemotron™ 3 Nano Omni is a 30B-A3B open multimodal model designed to function as a perception and context sub-agent in enterprise agent systems. It accepts text, image, video, and...
NVIDIA Nemotron 3 Super is a 120B-parameter open hybrid MoE model, activating just 12B parameters for maximum compute efficiency and accuracy in complex multi-agent applications. Built on a hybrid Mamba-Transformer...
NVIDIA Nemotron 3 Ultra is an open frontier-reasoning and orchestration model from NVIDIA, with 55B active parameters out of 550B total (MoE). Built on a hybrid Transformer-Mamba mixture-of-experts architecture, it...
NVIDIA Nemotron 3 Ultra is an open frontier-reasoning and orchestration model from NVIDIA, with 55B active parameters out of 550B total (MoE). Built on a hybrid Transformer-Mamba mixture-of-experts architecture, it...
Switchyard is an open-source model routing system that switches between multiple models to optimize the cost of requests. By default it will use OpenRouter market data to select the most...
GPT-3.5 Turbo is OpenAI's fastest model. It can understand and generate natural language or code, and is optimized for chat and traditional completion tasks. Training data up to Sep 2021.
This model offers four times the context length of gpt-3.5-turbo, allowing it to support approximately 20 pages of text in a single request at a higher cost. Training data: up...
OpenAI's flagship model, GPT-4 is a large-scale multimodal language model capable of solving difficult problems with greater accuracy than previous models due to its broader general knowledge and advanced reasoning...
GPT-4.1 is a flagship large language model optimized for advanced instruction following, real-world software engineering, and long-context reasoning. It supports a 1 million token context window and outperforms GPT-4o and...
GPT-4.1 is a flagship large language model optimized for advanced instruction following, real-world software engineering, and long-context reasoning. It supports a 1 million token context window and outperforms GPT-4o and...
GPT-4.1 Mini is a mid-sized model delivering performance competitive with GPT-4o at substantially lower latency and cost. It retains a 1 million token context window and scores 45.1% on hard...
GPT-4.1 Mini is a mid-sized model delivering performance competitive with GPT-4o at substantially lower latency and cost. It retains a 1 million token context window and scores 45.1% on hard...
For tasks that demand low latency, GPT‑4.1 nano is the fastest and cheapest model in the GPT-4.1 series. It delivers exceptional performance at a small size with its 1 million...
For tasks that demand low latency, GPT‑4.1 nano is the fastest and cheapest model in the GPT-4.1 series. It delivers exceptional performance at a small size with its 1 million...
The latest GPT-4 Turbo model with vision capabilities. Vision requests can now use JSON mode and function calling. Training data: up to December 2023.
GPT-4o ("o" for "omni") is OpenAI's latest AI model, supporting both text and image inputs with text outputs. It maintains the intelligence level of [GPT-4 Turbo](/models/openai/gpt-4-turbo) while being twice as...
GPT-4o ("o" for "omni") is OpenAI's latest AI model, supporting both text and image inputs with text outputs. It maintains the intelligence level of [GPT-4 Turbo](/models/openai/gpt-4-turbo) while being twice as...
The 2024-08-06 version of GPT-4o offers improved performance in structured outputs, with the ability to supply a JSON schema in the respone_format. Read more [here](https://openai.com/index/introducing-structured-outputs-in-the-api/). GPT-4o ("o" for "omni") is...
The 2024-11-20 version of GPT-4o offers a leveled-up creative writing ability with more natural, engaging, and tailored writing to improve relevance & readability. It’s also better at working with uploaded...
GPT-4o ("o" for "omni") is OpenAI's latest AI model, supporting both text and image inputs with text outputs. It maintains the intelligence level of [GPT-4 Turbo](/models/openai/gpt-4-turbo) while being twice as...
GPT-4o mini is OpenAI's newest model after [GPT-4 Omni](/models/openai/gpt-4o), supporting both text and image inputs with text outputs. As their most advanced small model, it is many multiples more affordable...
GPT-4o mini is OpenAI's newest model after [GPT-4 Omni](/models/openai/gpt-4o), supporting both text and image inputs with text outputs. As their most advanced small model, it is many multiples more affordable...
GPT-4o mini is OpenAI's newest model after [GPT-4 Omni](/models/openai/gpt-4o), supporting both text and image inputs with text outputs. As their most advanced small model, it is many multiples more affordable...
GPT-5 is OpenAI’s most advanced model, offering major improvements in reasoning, code quality, and user experience. It is optimized for complex tasks that require step-by-step reasoning, instruction following, and accuracy...
GPT-5.1 is the latest frontier-grade model in the GPT-5 series, offering stronger general-purpose reasoning, improved instruction adherence, and a more natural conversational style compared to GPT-5. It uses adaptive reasoning...
GPT-5.1 is the latest frontier-grade model in the GPT-5 series, offering stronger general-purpose reasoning, improved instruction adherence, and a more natural conversational style compared to GPT-5. It uses adaptive reasoning...
GPT-5.1-Codex is a specialized version of GPT-5.1 optimized for software engineering and coding workflows. It is designed for both interactive development sessions and long, independent execution of complex engineering tasks....
GPT-5.1-Codex-Max is OpenAI’s latest agentic coding model, designed for long-running, high-context software development tasks. It is based on an updated version of the 5.1 reasoning stack and trained on agentic...
GPT-5.1-Codex-Mini is a smaller and faster version of GPT-5.1-Codex
GPT-5.2 is the latest frontier-grade model in the GPT-5 series, offering stronger agentic and long context perfomance compared to GPT-5.1. It uses adaptive reasoning to allocate computation dynamically, responding quickly...
GPT-5.2 is the latest frontier-grade model in the GPT-5 series, offering stronger agentic and long context perfomance compared to GPT-5.1. It uses adaptive reasoning to allocate computation dynamically, responding quickly...
GPT-5.2 Chat (AKA Instant) is the fast, lightweight member of the 5.2 family, optimized for low-latency chat while retaining strong general intelligence. It uses adaptive reasoning to selectively “think” on...
GPT-5.2-Codex is an upgraded version of GPT-5.1-Codex optimized for software engineering and coding workflows. It is designed for both interactive development sessions and long, independent execution of complex engineering tasks....
GPT-5.2 Pro is OpenAI’s most advanced model, offering major improvements in agentic coding and long context performance over GPT-5 Pro. It is optimized for complex tasks that require step-by-step reasoning,...
GPT-5.2 Pro is OpenAI’s most advanced model, offering major improvements in agentic coding and long context performance over GPT-5 Pro. It is optimized for complex tasks that require step-by-step reasoning,...
GPT-5.3-Codex is OpenAI’s most advanced agentic coding model, combining the frontier software engineering performance of GPT-5.2-Codex with the broader reasoning and professional knowledge capabilities of GPT-5.2. It achieves state-of-the-art results...
GPT-5.4 is OpenAI’s latest frontier model, unifying the Codex and GPT lines into a single system. It features a 1M+ token context window (922K input, 128K output) with support for...
GPT-5.4 is OpenAI’s latest frontier model, unifying the Codex and GPT lines into a single system. It features a 1M+ token context window (922K input, 128K output) with support for...
[GPT-5.4](https://openrouter.ai/openai/gpt-5.4) Image 2 combines OpenAI's GPT-5.4 model with state-of-the-art image generation capabilities from GPT Image 2. It enables rich multimodal workflows, allowing users to seamlessly move between reasoning, coding, and...
GPT-5.4 mini brings the core capabilities of GPT-5.4 to a faster, more efficient model optimized for high-throughput workloads. It supports text and image inputs with strong performance across reasoning, coding,...
GPT-5.4 mini brings the core capabilities of GPT-5.4 to a faster, more efficient model optimized for high-throughput workloads. It supports text and image inputs with strong performance across reasoning, coding,...
GPT-5.4 nano is the most lightweight and cost-efficient variant of the GPT-5.4 family, optimized for speed-critical and high-volume tasks. It supports text and image inputs and is designed for low-latency...
GPT-5.4 nano is the most lightweight and cost-efficient variant of the GPT-5.4 family, optimized for speed-critical and high-volume tasks. It supports text and image inputs and is designed for low-latency...
GPT-5.4 Pro is OpenAI's most advanced model, building on GPT-5.4's unified architecture with enhanced reasoning capabilities for complex, high-stakes tasks. It features a 1M+ token context window (922K input, 128K...
GPT-5.4 Pro is OpenAI's most advanced model, building on GPT-5.4's unified architecture with enhanced reasoning capabilities for complex, high-stakes tasks. It features a 1M+ token context window (922K input, 128K...
GPT-5.5 is OpenAI’s frontier model designed for complex professional workloads, building on GPT-5.4 with stronger reasoning, higher reliability, and improved token efficiency on hard tasks. It features a 1M+ token...
GPT-5.5 is OpenAI’s frontier model designed for complex professional workloads, building on GPT-5.4 with stronger reasoning, higher reliability, and improved token efficiency on hard tasks. It features a 1M+ token...
GPT-5.5 Pro is OpenAI’s high-capability model optimized for deep reasoning and accuracy on complex, high-stakes workloads. It features a 1M+ token context window (922K input, 128K output) with support for...
GPT-5.5 Pro is OpenAI’s high-capability model optimized for deep reasoning and accuracy on complex, high-stakes workloads. It features a 1M+ token context window (922K input, 128K output) with support for...
GPT-5.6 Luna is a fast, cost-efficient model in OpenAI's GPT-5.6 series. It is suited for high-volume, latency-sensitive tasks such as chat, classification, and lightweight agentic workflows, providing capable reasoning for...
GPT-5.6 Luna is a fast, cost-efficient model in OpenAI's GPT-5.6 series. It is suited for high-volume, latency-sensitive tasks such as chat, classification, and lightweight agentic workflows, providing capable reasoning for...
GPT-5.6 Luna Pro is the same underlying model as [GPT-5.6 Luna](https://openrouter.ai/openai/gpt-5.6-luna), served with `reasoning.mode` set to `pro` for higher-quality responses on complex tasks. **Cost note:** pro mode spends far more...
GPT-5.6 Luna Pro is the same underlying model as [GPT-5.6 Luna](https://openrouter.ai/openai/gpt-5.6-luna), served with `reasoning.mode` set to `pro` for higher-quality responses on complex tasks. **Cost note:** pro mode spends far more...
GPT-5.6 Sol is the flagship model in OpenAI's GPT-5.6 series. It is suited for complex reasoning, coding, and agentic workflows, and is particularly strong at command-line and multi-step coding tasks...
GPT-5.6 Sol is the flagship model in OpenAI's GPT-5.6 series. It is suited for complex reasoning, coding, and agentic workflows, and is particularly strong at command-line and multi-step coding tasks...
GPT-5.6 Sol Pro is the same underlying model as [GPT-5.6 Sol](https://openrouter.ai/openai/gpt-5.6-sol), served with `reasoning.mode` set to `pro` for higher-quality responses on complex tasks. **Cost note:** pro mode spends far more...
GPT-5.6 Sol Pro is the same underlying model as [GPT-5.6 Sol](https://openrouter.ai/openai/gpt-5.6-sol), served with `reasoning.mode` set to `pro` for higher-quality responses on complex tasks. **Cost note:** pro mode spends far more...
GPT-5.6 Terra is a balanced model in OpenAI's GPT-5.6 series, positioned between the flagship Sol tier and the cost-efficient Luna tier. It is suited for everyday coding, reasoning, and agentic...
GPT-5.6 Terra is a balanced model in OpenAI's GPT-5.6 series, positioned between the flagship Sol tier and the cost-efficient Luna tier. It is suited for everyday coding, reasoning, and agentic...
GPT-5.6 Terra Pro is the same underlying model as [GPT-5.6 Terra](https://openrouter.ai/openai/gpt-5.6-terra), served with `reasoning.mode` set to `pro` for higher-quality responses on complex tasks. **Cost note:** pro mode spends far more...
GPT-5.6 Terra Pro is the same underlying model as [GPT-5.6 Terra](https://openrouter.ai/openai/gpt-5.6-terra), served with `reasoning.mode` set to `pro` for higher-quality responses on complex tasks. **Cost note:** pro mode spends far more...
GPT-5 is OpenAI’s most advanced model, offering major improvements in reasoning, code quality, and user experience. It is optimized for complex tasks that require step-by-step reasoning, instruction following, and accuracy...
[GPT-5](https://openrouter.ai/openai/gpt-5) Image combines OpenAI's GPT-5 model with state-of-the-art image generation capabilities. It offers major improvements in reasoning, code quality, and user experience while incorporating GPT Image 1's superior instruction following,...
GPT-5 Image Mini combines OpenAI's advanced language capabilities, powered by [GPT-5 Mini](https://openrouter.ai/openai/gpt-5-mini), with GPT Image 1 Mini for efficient image generation. This natively multimodal model features superior instruction following, text...
GPT-5 Mini is a compact version of GPT-5, designed to handle lighter-weight reasoning tasks. It provides the same instruction-following and safety-tuning benefits as GPT-5, but with reduced latency and cost....
GPT-5 Mini is a compact version of GPT-5, designed to handle lighter-weight reasoning tasks. It provides the same instruction-following and safety-tuning benefits as GPT-5, but with reduced latency and cost....
GPT-5-Nano is the smallest and fastest variant in the GPT-5 system, optimized for developer tools, rapid interactions, and ultra-low latency environments. While limited in reasoning depth compared to its larger...
GPT-5-Nano is the smallest and fastest variant in the GPT-5 system, optimized for developer tools, rapid interactions, and ultra-low latency environments. While limited in reasoning depth compared to its larger...
GPT-5 Pro is OpenAI’s most advanced model, offering major improvements in reasoning, code quality, and user experience. It is optimized for complex tasks that require step-by-step reasoning, instruction following, and...
GPT-5 Pro is OpenAI’s most advanced model, offering major improvements in reasoning, code quality, and user experience. It is optimized for complex tasks that require step-by-step reasoning, instruction following, and...
GPT-6.1 Sol is an upgrade to GPT-6 Sol from OpenAI, positioned below the flagship GPT-6 Astra in the GPT-6 series. It is suited for agentic coding, computer use, document-heavy professional...
GPT-6.1 Sol is an upgrade to GPT-6 Sol from OpenAI, positioned below the flagship GPT-6 Astra in the GPT-6 series. It is suited for agentic coding, computer use, document-heavy professional...
GPT-6.1 Sol Pro is the same underlying model as [GPT-6.1 Sol](https://openrouter.ai/openai/gpt-6.1-sol), served with `reasoning.mode` set to `pro` for higher-quality responses on complex tasks. **Cost note:** pro mode spends far more...
GPT-6.1 Sol Pro is the same underlying model as [GPT-6.1 Sol](https://openrouter.ai/openai/gpt-6.1-sol), served with `reasoning.mode` set to `pro` for higher-quality responses on complex tasks. **Cost note:** pro mode spends far more...
GPT-6.1 Terra is the mid-tier model in OpenAI's GPT-6.1 series, priced below GPT-6.1 Sol. It is suited for everyday agentic, coding, and writing work, is particularly strong at understanding screenshots...
GPT-6.1 Terra is the mid-tier model in OpenAI's GPT-6.1 series, priced below GPT-6.1 Sol. It is suited for everyday agentic, coding, and writing work, is particularly strong at understanding screenshots...
GPT-6.1 Terra Pro is the same underlying model as [GPT-6.1 Terra](https://openrouter.ai/openai/gpt-6.1-terra), served with `reasoning.mode` set to `pro` for higher-quality responses on complex tasks. **Cost note:** pro mode spends far more...
GPT-6.1 Terra Pro is the same underlying model as [GPT-6.1 Terra](https://openrouter.ai/openai/gpt-6.1-terra), served with `reasoning.mode` set to `pro` for higher-quality responses on complex tasks. **Cost note:** pro mode spends far more...
GPT-6.2 Astra is the flagship model in the GPT-6.2 series from OpenAI, built for the hardest projects. It is suited for real-world engineering, professional work and scientific research, with stronger...
GPT-6.2 Astra is the flagship model in the GPT-6.2 series from OpenAI, built for the hardest projects. It is suited for real-world engineering, professional work and scientific research, with stronger...
GPT-6.2 Astra Pro is the same underlying model as [GPT-6.2 Astra](https://openrouter.ai/openai/gpt-6.2-astra), served with `reasoning.mode` set to `pro` for higher-quality responses on complex tasks. **Cost note:** pro mode spends far more...
GPT-6.2 Astra Pro is the same underlying model as [GPT-6.2 Astra](https://openrouter.ai/openai/gpt-6.2-astra), served with `reasoning.mode` set to `pro` for higher-quality responses on complex tasks. **Cost note:** pro mode spends far more...
GPT-6.2 Sol is an upgrade to GPT-6.1 Sol from OpenAI, positioned below the flagship GPT-6.2 Astra in the GPT-6.2 series. It is an efficient, versatile model for everyday work, suited...
GPT-6.2 Sol is an upgrade to GPT-6.1 Sol from OpenAI, positioned below the flagship GPT-6.2 Astra in the GPT-6.2 series. It is an efficient, versatile model for everyday work, suited...
GPT-6.2 Sol Pro is the same underlying model as [GPT-6.2 Sol](https://openrouter.ai/openai/gpt-6.2-sol), served with `reasoning.mode` set to `pro` for higher-quality responses on complex tasks. **Cost note:** pro mode spends far more...
GPT-6.2 Sol Pro is the same underlying model as [GPT-6.2 Sol](https://openrouter.ai/openai/gpt-6.2-sol), served with `reasoning.mode` set to `pro` for higher-quality responses on complex tasks. **Cost note:** pro mode spends far more...
GPT-6 Astra is OpenAI's flagship model for demanding end-to-end work. It is suited for advanced analysis, software engineering, deep research, scientific work, and document creation, with particular strengths in long-horizon...
GPT-6 Astra is OpenAI's flagship model for demanding end-to-end work. It is suited for advanced analysis, software engineering, deep research, scientific work, and document creation, with particular strengths in long-horizon...
GPT-6 Astra Pro is the same underlying model as [GPT-6 Astra](https://openrouter.ai/openai/gpt-6-astra), served with `reasoning.mode` set to `pro` for higher-quality responses on complex tasks. **Cost note:** pro mode spends far more...
GPT-6 Astra Pro is the same underlying model as [GPT-6 Astra](https://openrouter.ai/openai/gpt-6-astra), served with `reasoning.mode` set to `pro` for higher-quality responses on complex tasks. **Cost note:** pro mode spends far more...
GPT-6 Luna is the fast, cost-efficient model in OpenAI's GPT-6 series, positioned below GPT-6 Sol. It is suited for high-volume and latency-sensitive workloads such as chat, classification, and lightweight agentic...
GPT-6 Luna is the fast, cost-efficient model in OpenAI's GPT-6 series, positioned below GPT-6 Sol. It is suited for high-volume and latency-sensitive workloads such as chat, classification, and lightweight agentic...
GPT-6 Luna Pro is the same underlying model as [GPT-6 Luna](https://openrouter.ai/openai/gpt-6-luna), served with `reasoning.mode` set to `pro` for higher-quality responses on complex tasks. **Cost note:** pro mode spends far more...
GPT-6 Luna Pro is the same underlying model as [GPT-6 Luna](https://openrouter.ai/openai/gpt-6-luna), served with `reasoning.mode` set to `pro` for higher-quality responses on complex tasks. **Cost note:** pro mode spends far more...
GPT-6 Sol is the cost-efficient high-end model in OpenAI's GPT-6 series, positioned below the flagship GPT-6 Astra and above the fast GPT-6 Luna tier. It is suited for demanding professional...
GPT-6 Sol is the cost-efficient high-end model in OpenAI's GPT-6 series, positioned below the flagship GPT-6 Astra and above the fast GPT-6 Luna tier. It is suited for demanding professional...
GPT-6 Sol Pro is the same underlying model as [GPT-6 Sol](https://openrouter.ai/openai/gpt-6-sol), served with `reasoning.mode` set to `pro` for higher-quality responses on complex tasks. **Cost note:** pro mode spends far more...
GPT-6 Sol Pro is the same underlying model as [GPT-6 Sol](https://openrouter.ai/openai/gpt-6-sol), served with `reasoning.mode` set to `pro` for higher-quality responses on complex tasks. **Cost note:** pro mode spends far more...
This model always redirects to the latest model in the GPT Astra family.
The gpt-audio model is OpenAI's first generally available audio model. The new snapshot features an upgraded decoder for more natural sounding voices and maintains better voice consistency. Audio is priced...
A cost-efficient version of GPT Audio. The new snapshot features an upgraded decoder for more natural sounding voices and maintains better voice consistency. Input is priced at $0.60 per million...
GPT Chat Latest points to OpenAI's stable API alias `chat-latest` that always resolves to the latest Instant chat model used in ChatGPT. As OpenAI rolls out new Instant model updates...
This model always redirects to the latest model in the GPT Luna family.
This model always redirects to the latest model in the GPT Mini family.
gpt-oss-120b is an open-weight, 117B-parameter Mixture-of-Experts (MoE) language model from OpenAI designed for high-reasoning, agentic, and general-purpose production use cases. It activates 5.1B parameters per forward pass and is optimized...
gpt-oss-120b is an open-weight, 117B-parameter Mixture-of-Experts (MoE) language model from OpenAI designed for high-reasoning, agentic, and general-purpose production use cases. It activates 5.1B parameters per forward pass and is optimized...
gpt-oss-20b is an open-weight 21B parameter model released by OpenAI under the Apache 2.0 license. It uses a Mixture-of-Experts (MoE) architecture with 3.6B active parameters per forward pass, optimized for...
gpt-oss-20b is an open-weight 21B parameter model released by OpenAI under the Apache 2.0 license. It uses a Mixture-of-Experts (MoE) architecture with 3.6B active parameters per forward pass, optimized for...
gpt-oss-safeguard-20b is a safety reasoning model from OpenAI built upon gpt-oss-20b. This open-weight, 21B-parameter Mixture-of-Experts (MoE) model offers lower latency for safety tasks like content classification, LLM filtering, and trust...
This model always redirects to the latest model in the GPT Sol family.
This model always redirects to the latest model in the GPT Terra family.
o3 is a well-rounded and powerful model across domains. It sets a new standard for math, science, coding, and visual reasoning tasks. It also excels at technical writing and instruction-following....
o3 is a well-rounded and powerful model across domains. It sets a new standard for math, science, coding, and visual reasoning tasks. It also excels at technical writing and instruction-following....
OpenAI o3-mini is a cost-efficient language model optimized for STEM reasoning tasks, particularly excelling in science, mathematics, and coding. This model supports the `reasoning_effort` parameter, which can be set to...
The o-series of models are trained with reinforcement learning to think before they answer and perform complex reasoning. The o3-pro model uses more compute to think harder and provide consistently...
OpenAI o4-mini is a compact reasoning model in the o-series, optimized for fast, cost-efficient performance while retaining strong multimodal and agentic capabilities. It supports tool use and demonstrates competitive reasoning...
Auto is OpenRouter’s flagship model router, built for general-purpose usage. It adapts to each request, choosing cost-optimal models and adjusting reasoning effort. It balances task difficulty, benchmark results, live prices,...
The experimental version of our Auto Router where we test new improvements. Use it to get the latest and greatest version of our general purpose auto router, but expect beta...
Auto Code is OpenRouter’s router built specifically for coding agents and harnesses. It uses the same routing engine as the [Auto Router](https://openrouter.ai/openrouter/auto), with every request treated as part of an...
Transform your natural language requests into structured OpenRouter API request objects. Describe what you want to accomplish with AI models, and Body Builder will construct the appropriate API calls. Example:...
The simplest way to get free inference. openrouter/free is a router that selects free models at random from the models available on OpenRouter. The router smartly filters for models that...
Fusion turns your prompt into a small multi-model deliberation. A panel of expert models (see below) analyzes your prompt in parallel with web search and web fetch enabled, then a...
The Pareto Router maintains a tiered shortlist of strong coding models, ranked by [Artificial Analysis](https://artificialanalysis.ai/) coding percentiles. Set min_coding_score between 0 and 1 on the [pareto-router plugin](https://openrouter.ai/docs/guides/routing/routers/pareto-router#the-min_coding_score-parameter) to control how...
Particle-1 is a token-efficient agentic coding model from Particle AI, post-trained on DeepSeek V4.1 Flash. It is suited for software engineering work such as terminal workflows, codebase changes, and data...
Particle-1 is a token-efficient agentic coding model from Particle AI, post-trained on DeepSeek V4.1 Flash. It is suited for software engineering work such as terminal workflows, codebase changes, and data...
Perceptron Mk1 (Mark One) is Perceptron's highest-quality vision-language model for video and embodied reasoning.** It accepts image and video inputs paired with natural language queries, and produces detailed visual understanding...
Perceptron Mk1.5 is Perceptron's embodied reasoning model for physical agents. It accepts text, image, video, and audio input, and answers with text plus optional structured annotations: points, boxes, polygons, tracks,...
Sonar is lightweight, affordable, fast, and simple to use — now featuring citations and the ability to customize sources. It is designed for companies seeking to integrate lightweight question-and-answer features...
Sonar Deep Research is a research-focused model designed for multi-step retrieval, synthesis, and reasoning across complex topics. It autonomously searches, reads, and evaluates sources, refining its approach as it gathers...
Note: Sonar Pro pricing includes Perplexity search pricing. See [details here](https://docs.perplexity.ai/guides/pricing#detailed-pricing-breakdown-for-sonar-reasoning-pro-and-sonar-pro) For enterprises seeking more advanced capabilities, the Sonar Pro API can handle in-depth, multi-step queries with added extensibility, like...
Exclusively available on the OpenRouter API, Sonar Pro's new Pro Search mode is Perplexity's most advanced agentic search system. It is designed for deeper reasoning and analysis. Pricing is based...
Note: Sonar Pro pricing includes Perplexity search pricing. See [details here](https://docs.perplexity.ai/guides/pricing#detailed-pricing-breakdown-for-sonar-reasoning-pro-and-sonar-pro) Sonar Reasoning Pro is a premier reasoning model powered by DeepSeek R1 with Chain of Thought (CoT). Designed for...
Bonsai 2 27B is a 27B-parameter reasoning model from PrismML derived from Qwen3.8-27B. It supports coding, mathematics, tool calling, and image understanding with a 262K-token context window. Ternary compression shrinks...
Ternary Bonsai 2 9B is PrismML’s compact, open-weight language model for conversational AI, reasoning, and tool use. Built on Qwen3.5, it uses ternary weights packed into 2 bits for efficient...
Qwen2.5 72B is the latest series of Qwen large language models. Qwen2.5 brings the following improvements upon Qwen2: - Significantly more knowledge and has greatly improved capabilities in coding and...
Qwen2.5-Coder is the latest series of Code-Specific Qwen large language models (formerly known as CodeQwen). Qwen2.5-Coder brings the following improvements upon CodeQwen1.5: - Significantly improvements in **code generation**, **code reasoning**...
Qwen-Plus, based on the Qwen2.5 foundation model, is a 131K context model with a balanced performance, speed, and cost combination.
Qwen2.5-VL is proficient in recognizing common objects such as flowers, birds, fish, and insects. It is also highly capable of analyzing texts, charts, icons, graphics, and layouts within images.
Qwen3-14B is a dense 14.8B parameter causal language model from the Qwen3 series, designed for both complex reasoning and efficient dialogue. It supports seamless switching between a "thinking" mode for...
Qwen3-235B-A22B-Instruct-2507 is a multilingual, instruction-tuned mixture-of-experts language model based on the Qwen3-235B architecture, with 22B active parameters per forward pass. It is optimized for general-purpose text generation, including instruction following,...
Qwen3-235B-A22B-Thinking-2507 is a high-performance, open-weight Mixture-of-Experts (MoE) language model optimized for complex reasoning tasks. It activates 22B of its 235B parameters per forward pass and natively supports up to 262,144...
Qwen3, the latest generation in the Qwen large language model series, features both dense and mixture-of-experts (MoE) architectures to excel in reasoning, multilingual support, and advanced agent tasks. Its unique...
Qwen3-30B-A3B-Instruct-2507 is a 30.5B-parameter mixture-of-experts language model from Qwen, with 3.3B active parameters per inference. It operates in non-thinking mode and is designed for high-quality instruction following, multilingual understanding, and...
Qwen3-32B is a dense 32.8B parameter causal language model from the Qwen3 series, optimized for both complex reasoning and efficient dialogue. It supports seamless switching between a "thinking" mode for...
The Qwen3.5 122B-A10B native vision-language model is built on a hybrid architecture that integrates a linear attention mechanism with a sparse mixture-of-experts model, achieving higher inference efficiency. In terms of...
The Qwen3.5 27B native vision-language Dense model incorporates a linear attention mechanism, delivering fast response times while balancing inference speed and performance. Its overall capabilities are comparable to those of...
The Qwen3.5 Series 35B-A3B is a native vision-language model designed with a hybrid architecture that integrates linear attention mechanisms and a sparse mixture-of-experts model, achieving higher inference efficiency. Its overall...
The Qwen3.5 series 397B-A17B native vision-language model is built on a hybrid architecture that integrates a linear attention mechanism with a sparse mixture-of-experts model, achieving higher inference efficiency. It delivers...
Qwen3.5-9B is a multimodal foundation model from the Qwen3.5 family, designed to deliver strong reasoning, coding, and visual understanding in an efficient 9B-parameter architecture. It uses a unified vision-language design...
The Qwen3.5 native vision-language Flash models are built on a hybrid architecture that integrates a linear attention mechanism with a sparse mixture-of-experts model, achieving higher inference efficiency. Compared to the...
The Qwen3.5 native vision-language series Plus models are built on a hybrid architecture that integrates linear attention mechanisms with sparse mixture-of-experts models, achieving higher inference efficiency. In a variety of...
Qwen3.5 Plus (April 2026) is a large-scale multimodal language model from Alibaba. It accepts text, image, and video input and produces text output, with a 1M token context window. This...
Qwen3.6 27B is a dense 27-billion-parameter language model from the Qwen Team at Alibaba, released in April 2026. It features hybrid multimodal capabilities — accepting text, image, and video inputs...
Qwen3.6-35B-A3B is an open-weight multimodal model from Alibaba Cloud with 35 billion total parameters and 3 billion active parameters per token. It uses a hybrid sparse mixture-of-experts architecture combining Gated...
Qwen3.6 Flash is a fast, efficient language model from Alibaba's Qwen 3.6 series. It supports text, image, and video input with a 1M token context window. Tiered pricing kicks in...
Qwen 3.6 Plus builds on a hybrid architecture that combines efficient linear attention with sparse mixture-of-experts routing, enabling strong scalability and high-performance inference. Compared to the 3.5 series, it delivers...
Qwen3.7 Flash is a vision-language reasoning model from Alibaba. It is suited for multimodal agents, visual coding, search, and computer interaction, with strengths in object recognition, spatial understanding, and real-world...
Qwen3.7-Max is the flagship model in Alibaba's Qwen3.7 series. It supports text input and output and is designed for agent-centric workloads, with particular strengths in coding, office and productivity tasks,...
Qwen3.7-Plus is a cost-effective model in Alibaba's Qwen3.7 series. It supports text and image input with text output, building on the series' text capabilities with a comprehensive upgrade to its...
Qwen3.7-Plus is a cost-effective model in Alibaba's Qwen3.7 series. It supports text and image input with text output, building on the series' text capabilities with a comprehensive upgrade to its...
Qwen3.8 2.4T A95B is an open-weight sparse mixture-of-experts model from Qwen and the open-weight variant of [Qwen3.8 Max](/qwen/qwen3.8-max), with 95 billion active parameters out of 2.4 trillion total. It is...
Qwen3.8 27B is an open-weight dense vision-language model from Qwen. It is suited for coding, professional workflows, research, multimodal interaction, and long-running agent tasks, with flexible thinking that can be...
Qwen3.8 27B is an open-weight dense vision-language model from Qwen. It is suited for coding, professional workflows, research, multimodal interaction, and long-running agent tasks, with flexible thinking that can be...
Qwen3.8 Flash is a multimodal reasoning model from Alibaba. It is suited for coding assistance, agentic workflows, visual understanding, document and codebase analysis, desktop interaction, chart analysis, and long-video analysis.
Qwen3.8 Max 0902 is an updated snapshot of Qwen3.8 Max from Alibaba's Qwen team. It is a 2.4-trillion-parameter mixture-of-experts model that accepts text, image, and video input and returns text,...
Qwen3.8 Max 0902 is an updated snapshot of Qwen3.8 Max from Alibaba's Qwen team. It is a 2.4-trillion-parameter mixture-of-experts model that accepts text, image, and video input and returns text,...
Qwen3.8 Max Prime is a higher-throughput variant of Qwen3.8 Max from Alibaba's Qwen team, served as a separate SKU at a higher price point. It accepts text, image, and video...
Qwen3.8 Omni Flash is an omni-modal reasoning model from Alibaba, the first Qwen model built around agentic capabilities with native audio-video understanding. It is suited for audio-video analysis and summarization,...
Qwen3-Coder-480B-A35B-Instruct is a Mixture-of-Experts (MoE) code generation model developed by the Qwen team. It is optimized for agentic coding tasks such as function calling, tool use, and long-context reasoning over...
Qwen3-Coder-30B-A3B-Instruct is a 30.5B parameter Mixture-of-Experts (MoE) model with 128 experts (8 active per forward pass), designed for advanced code generation, repository-scale understanding, and agentic tool use. Built on the...
Qwen3 Coder Flash is Alibaba's fast and cost efficient version of their proprietary Qwen3 Coder Plus. It is a powerful coding agent model specializing in autonomous programming via tool calling...
Qwen3-Coder-Next is an open-weight causal language model optimized for coding agents and local development workflows. It uses a sparse MoE design with 80B total parameters and only 3B activated per...
Qwen3-Next-80B-A3B-Instruct is an instruction-tuned chat model in the Qwen3-Next series optimized for fast, stable responses without “thinking” traces. It targets complex tasks across reasoning, code generation, knowledge QA, and multilingual...
Qwen3-VL-235B-A22B Instruct is an open-weight multimodal model that unifies strong text generation with visual understanding across images and video. The Instruct model targets general vision-language use (VQA, document parsing, chart/table...
Qwen3-VL-30B-A3B-Instruct is a multimodal model that unifies strong text generation with visual understanding for images and videos. Its Instruct variant optimizes instruction-following for general multimodal tasks. It excels in perception...
Qwen3-VL-8B-Instruct is a multimodal vision-language model from the Qwen3-VL series, built for high-fidelity understanding and reasoning across text, images, and video. It features improved multimodal fusion with Interleaved-MRoPE for long-horizon...
Qwen4 Max is the flagship model in Alibaba's Qwen4 series. It accepts text, image, and video input and returns text, with a 1M-token context window and reasoning enabled by default....
DaisyRose 27B is an open-weight multimodal reasoning model from RadixArk, trained from [Qwen3.8 27B](https://openrouter.ai/qwen/qwen3.8-27b) using reinforcement learning only on long-horizon agent tasks, with no supervised fine-tuning or distillation. It is...
Flint is an open-weight mixture-of-experts model from Recursive, with about 2.5B active parameters out of 40B total, built by Recursive's autoresearch system Eureka Machine for coding agents. It is suited...
Beam 1.0 is Reflection's open-weight reasoning model for coding and long-horizon agentic workflows. It uses a sparse mixture-of-experts architecture with 501B total and 23B active parameters per token, and a...
Reka Edge is an extremely efficient 7B multimodal vision-language model that accepts image/video+text inputs and generates text outputs. This model is optimized specifically to deliver industry-leading performance in image understanding,...
Reka Flash 3 is a general-purpose, instruction-tuned large language model with 21 billion parameters, developed by Reka. It excels at general chat, coding tasks, instruction-following, and function calling. Featuring a...
Relace Apply 3 is a specialized code-patching LLM that merges AI-suggested edits straight into your source files. It can apply updates from GPT-4o, Claude, and others into your files at...
The relace-search model uses 4-12 `view_file` and `grep` tools in parallel to explore a codebase and return relevant files to the user request. In contrast to RAG, relace-search performs agentic...
Ripple-65B-Preview by ricursiveAI, a 65B model with native tool calling, served in BF16.
Fugu Max is the cost-performance model in Sakana AI's Fugu family. Rather than a single monolithic model, Fugu is a learned multi-agent orchestration system: a language model trained to route...
Fugu Ultra is the higher-performance model in Sakana AI's Fugu family. Rather than a single monolithic model, Fugu is a learned multi-agent orchestration system: a language model trained to route...
Fugu Ultra v2 is the higher-performance model in Sakana AI's Fugu family. Rather than a single monolithic model, Fugu is a learned multi-agent orchestration system: a language model trained to...
Sakana Namazu is a Japanese-specialized reasoning model from Sakana AI, based on Kimi K2.6 with additional training for Japanese language and business contexts. It is suited for Japanese instruction following,...
Euryale L3.1 70B v2.2 is a model focused on creative roleplay from [Sao10k](https://ko-fi.com/sao10k). It is the successor of [Euryale L3 70B v2.1](/models/sao10k/l3-euryale-70b).
Lunaris 8B is a versatile generalist and roleplaying model based on Llama 3. It's a strategic merge of multiple models, designed to balance creativity with improved logic and general knowledge....
Extract is a description-driven entity extraction model from ScaleDown. Instead of a fixed set of entity types, callers describe the fields they want in plain English through `provider.options.scaledown.entities`, and the...
Summarize is an abstractive summarization model from ScaleDown. It condenses long text and documents into a fluent, faithful rewrite rather than lifting sentences from the source, and is suited for...
Step 3.7 Flash is StepFun's latest high-efficiency multimodal Mixture-of-Experts model. It pairs a 196B-parameter language backbone with a vision encoder for native image and video understanding, activating roughly 11B parameters...
Step 5 Preview is StepFun's flagship model for agentic work, built on a sparse Mixture-of-Experts architecture (27B active / 600B total parameters). It performs strongly in software engineering and professional...
Hy-MT2-1.8B is a compact 1.8B-parameter translation model from Tencent. It supports 33 language pairs and five Chinese dialect and minority-language pairs, with workflows for structured, delimiter-based, contextual, glossary-based, and style-guided...
Hy-MT2-30B-A3B is Tencent's flagship translation model in the Hy-MT2 family. It supports 33 language pairs and five Chinese dialect and minority-language pairs, with workflows for structured, delimiter-based, contextual, glossary-based, and...
Hy-MT2-7B is a 7B-parameter translation model from Tencent. It supports 33 language pairs and five Chinese dialect and minority-language pairs, with workflows for structured, delimiter-based, contextual, glossary-based, and style-guided translation.
Hy3 is a 295B-parameter Mixture-of-Experts model from Tencent (21B active, 192 experts with top-8 routing) built for reasoning, agentic workflows, and real-world production use. It supports a configurable reasoning effort:...
The official release of Hy4 delivers comprehensive and substantial improvements over Hy4 Preview, with particularly strong gains in agentic capabilities and performance across a wide range of long-horizon, real-world productivity...
Hy4 flash is a cost-efficient variant of [Hy4](https://openrouter.ai/tencent/hy4) from Tencent, served with roughly a quarter of the flagship's parameter count for lower latency and cost. It targets code assistance, agent...
Tencent: Hy4 preview is a mixture-of-experts model from Tencent, with 49B active parameters out of 770B total. It is designed for coding agents, complex tool-use workflows, and productivity tasks that...
Uncensored and creative writing model based on Mistral Small 3.2 24B with good recall, prompt adherence, and intelligence.
Skyfall 36B v2 is an enhanced iteration of Mistral Small 2501, specifically fine-tuned for improved creativity, nuanced writing, role-playing, and coherent storytelling.
UnslopNemo v4.1 is the latest addition from the creator of Rocinante, designed for adventure writing and role-play scenarios.
Inkling is an open-weight multimodal mixture-of-experts model from Thinking Machines Lab, with 41B active parameters out of 975B total. It is designed for general-purpose reasoning, coding, agentic and tool-use systems,...
Inkling 2 Mini is an open-weight multimodal looped transformer from Thinking Machines Lab, with 3.4B active parameters out of 26.4B total (MoE). Its middle layers run three times with shared...
Inkling 2 Small is an open-weight multimodal mixture-of-experts reasoning model from Thinking Machines Lab, with 17B active parameters out of 594B total. It is the first model in the second...
Inkling Small is an open-weight multimodal mixture-of-experts model from Thinking Machines Lab, with 12B active parameters out of 276B total. It is positioned as the smaller, more efficient member of...
Inkling Small is an open-weight multimodal mixture-of-experts model from Thinking Machines Lab, with 12B active parameters out of 276B total. It is positioned as the smaller, more efficient member of...
Jev Router picks the best model and reasoning effort for each request, balancing quality, speed, and cost. It runs on [Jev](https://openrouter.ai/~typesafe/jev-latest), TypeSafe's first System One model, and adapts as your...
Pareto is a multimodal composite model built for research, coding, and agentic workflows, while delivering frontier-level performance across a broad range of general-purpose tasks.
Pareto is a multimodal composite model built for research, coding, and agentic workflows, while delivering frontier-level performance across a broad range of general-purpose tasks. This is a preview of the...
Pareto is a multimodal composite model built for research, coding, and agentic workflows, while delivering frontier-level performance across a broad range of general-purpose tasks.
This model always redirects to the latest model in the Pareto family.
A recreation trial of the original MythoMax-L2-B13 but with updated models. #merge
Solar Mini 4 is Upstage's compact, cost-efficient language model, a 35B-parameter mixture-of-experts with 3B active parameters and a 524K context window. It is built for agentic use cases where response...
Solar Pro 4 is Upstage's cost-efficient large language model, featuring a 524K context window. It is built for long-horizon tasks and agentic workflows, with strong capabilities in office productivity, document-intensive...
Palmyra X5 is Writer's most advanced model, purpose-built for building and scaling AI agents across the enterprise. It delivers industry-leading speed and efficiency on context windows up to 1 million...
Grok 4.20 is a reasoning model from SpaceXSI with industry-leading speed and agentic tool calling capabilities. It combines the lowest hallucination rate on the market with strict prompt adherance, delivering...
Grok 4.20 Multi-Agent is a variant of SpaceXSI’s Grok 4.20 designed for collaborative, agent-based workflows. Multiple agents operate in parallel to conduct deep research, coordinate tool use, and synthesize information...
Grok 4.3 is a reasoning model from SpaceXSI. It accepts text and image inputs with text output, and is suited for agentic workflows, instruction-following tasks, and applications requiring high factual...
Grok 4.3 is a reasoning model from SpaceXSI. It accepts text and image inputs with text output, and is suited for agentic workflows, instruction-following tasks, and applications requiring high factual...
Grok 4.5 is a model from SpaceXSI with frontier performance on coding, knowledge work, and STEM.
Grok 4.6 is a model from SpaceXSI with frontier performance on coding, knowledge work, and STEM. It is succeeded by [Grok 4.7](/x-ai/grok-4.7).
Grok 4.6 is a model from SpaceXSI with frontier performance on coding, knowledge work, and STEM. It is succeeded by [Grok 4.7](/x-ai/grok-4.7).
Grok 4.7 is SpaceXSI's flagship model for coding, agentic tasks, and knowledge work, succeeding Grok 4.6. It is particularly strong at long-running software engineering tasks, verifying its own work, and...
Grok 4.7 is SpaceXSI's flagship model for coding, agentic tasks, and knowledge work, succeeding Grok 4.6. It is particularly strong at long-running software engineering tasks, verifying its own work, and...
This model always redirects to the latest Grok model from xAI.
MiMo-V2.5 is a native omnimodal model by Xiaomi. It delivers Pro-level agentic performance at roughly half the inference cost, while surpassing MiMo-V2-Omni in multimodal perception across image and video understanding...
MiMo-V2.5-Pro is Xiaomi’s flagship model, delivering strong performance in general agentic capabilities, complex software engineering, and long-horizon tasks, with top rankings on benchmarks such as ClawEval, GDPVal, and SWE-bench Pro....
MiMo-V2.6-Flash is an open-source foundation model developed by Xiaomi. Built on a Mixture-of-Experts architecture with 309B total parameters and 15B activated per token, it employs a hybrid attention mechanism for...
MiMo-V2.6-Pro is the flagship foundation model developed by Xiaomi. Built at a scale of over 1T parameters, it is designed to push the ceiling of capability for the most demanding...
MiMo-V2.6-Pro-UltraSpeed is the fast speed edition of Xiaomi's flagship foundation model, MiMo-V2.6-Pro. Built from the same 1T MiMo-V2.6-Pro checkpoint, it matches the original model in quality while delivering roughly 10x...
GLM-4.5 is our latest flagship foundation model, purpose-built for agent-based applications. It leverages a Mixture-of-Experts (MoE) architecture and supports a context length of up to 128k tokens. GLM-4.5 delivers significantly...
GLM-4.5-Air is the lightweight variant of our latest flagship model family, also purpose-built for agent-centric applications. Like GLM-4.5, it adopts the Mixture-of-Experts (MoE) architecture but with a more compact parameter...
GLM-4.5V is a vision-language foundation model for multimodal agent applications. Built on a Mixture-of-Experts (MoE) architecture with 106B parameters and 12B activated parameters, it achieves state-of-the-art results in video understanding,...
Compared with GLM-4.5, this generation brings several key improvements: Longer context window: The context window has been expanded from 128K to 200K tokens, enabling the model to handle more complex...
GLM-4.6V is a large multimodal model designed for high-fidelity visual understanding and long-context reasoning across images, documents, and mixed media. It supports up to 128K tokens, processes complex page layouts...
GLM-4.7 is Z.ai’s latest flagship model, featuring upgrades in two key areas: enhanced programming capabilities and more stable multi-step reasoning/execution. It demonstrates significant improvements in executing complex agent tasks while...
As a 30B-class SOTA model, GLM-4.7-Flash offers a new option that balances performance and efficiency. It is further optimized for agentic coding use cases, strengthening coding capabilities, long-horizon task planning,...
GLM-5 is Z.ai’s flagship open-source foundation model engineered for complex systems design and long-horizon agent workflows. Built for expert developers, it delivers production-grade performance on large-scale programming tasks, rivaling leading...
GLM-5.1 delivers a major leap in coding capability, with particularly significant gains in handling long-horizon tasks. Unlike previous models built around minute-level interactions, GLM-5.1 can work independently and continuously on...
GLM 5.2 is a large-scale reasoning model from Z.ai. It supports text input and output with a 1M-token context window, and is suited for long-horizon agent workflows, project-level software engineering,...
GLM 5.2 is a large-scale reasoning model from Z.ai. It supports text input and output with a 1M-token context window, and is suited for long-horizon agent workflows, project-level software engineering,...
GLM-5.3 is a large-scale reasoning model from Z.ai, built for complex software engineering and long-horizon agent tasks. It supports text input and output with a 1M-token context window, and improves...
GLM-5.3 is a large-scale reasoning model from Z.ai, built for complex software engineering and long-horizon agent tasks. It supports text input and output with a 1M-token context window, and improves...
GLM-5.3-Flash is a native multimodal model from Z.ai. It is suited for efficient coding and long-horizon agent tasks. Its hybrid sparse and linear attention architecture maintains accurate long-context behavior while...
GLM-5.3-Flash is a native multimodal model from Z.ai. It is suited for efficient coding and long-horizon agent tasks. Its hybrid sparse and linear attention architecture maintains accurate long-context behavior while...
GLM-5.3-FlashX is the high-speed variant of Z.ai's GLM-5.3-Flash, a native multimodal model delivering inference speeds of up to 200 tokens/s. Built on the same hybrid sparse and linear attention architecture...
GLM-5.3-FlashX-Legal (GLM-Legal) is Z.ai's legal-domain variant of GLM-5.3-FlashX, a native multimodal model that accepts text, image, and video input and produces text. It adds legal-domain mid-training and post-training with reinforcement...
GLM-5.3-Prime is the high-speed variant of Z.ai's GLM-5.3, inheriting its full capabilities while delivering 1.5–2× the output throughput through inference acceleration. It supports text input and output with a 1M-token...
GLM-5 Turbo is a new model from Z.ai designed for fast inference and strong performance in agent-driven environments such as OpenClaw scenarios. It is deeply optimized for real-world agent workflows...
GLM-5V-Turbo is Z.ai’s first native multimodal agent foundation model, built for vision-based coding and agent-driven tasks. It natively handles image, video, and text inputs, excels at long-horizon planning, complex coding,...
This model always redirects to the latest model in the GLM Flash family.
This model always redirects to the latest GLM model from Z.ai.