WaveSpeed for AI agents

WaveSpeed runs 1,143 image, video, audio and 3D generation models through one API, billed per run, and 137 language models through an OpenAI-compatible API, billed per token, with the same API key. This page says which model to pick for each job, what it costs, and how to call it. Prices are live quotes from the same calculator that bills the run; where a model is discounted, the discounted price is shown first. Prices quoted 2026-11-21 01:03 UTC.

How to call a model

Image

Generate images from a prompt

Product shots, portraits, scenes and marketing visuals from a text description.

  1. Seedream 5.5 (bytedance/seedream-v5.5, llms.txt): Seedream 5.5 Text to Image by ByteDance generates images from text prompts with improved photographic realism and aesthetics, native transparent-background output, aspect ratio selection, and 1K, 1.5K and 2K output tiers.
    Price: $0.019 per image at default settings (list price $0.02, 5% off)
  2. Seedream 5 Pro (bytedance/seedream-v5.0-pro, llms.txt): Seedream V5.0 Pro Text to Image by ByteDance generates high-quality images from text prompts, with aspect ratio selection, strong prompt following, and 1K / 2K output tiers for flexible image creation.
    Price: $0.0405 per image at default settings (list price $0.045, 10% off)
  3. GPT Image 2.6 (openai/gpt-image-2.6-flare/text-to-image, llms.txt): GPT Image 2.6 Flare Text-to-Image generates high-quality images from natural-language prompts with a focus on fast generation and rapid creative iteration, making it suitable for everyday image creation, concept development, marketing visuals, and…
    Price: $0.045 per image at default settings (list price $0.05, 10% off)
  4. Nano Banana 2.1 (google/nano-banana-2.1/text-to-image, llms.txt): Google Nano Banana 2.1 is the latest Nano Banana image generation model, with gains in visual design, subject consistency and prompt adherence, cleaner text rendering, 1K to 4K output and optional web search.
    Price: $0.036 per image at default settings (list price $0.04, 10% off)
  5. Nano Banana Pro (google/nano-banana-pro/text-to-image, llms.txt): Google's Nano Banana pro (Gemini 3.0 Pro Image) is a cutting-edge text-to-image model enabling high-res 4K image generation optimized for phones.
    Price: $0.126 per image at default settings (list price $0.14, 10% off)

Text, posters and typography

Posters, thumbnails, packaging and ads where the words must come out right.

  1. Nano Banana 2.1 (google/nano-banana-2.1/text-to-image, llms.txt): Gains in visual design and prompt adherence, with cleaner text rendering and 1K to 4K output.
    Price: $0.036 per image at default settings (list price $0.04, 10% off)
  2. GPT Image 2.6 Sunburst (openai/gpt-image-2.6-sunburst/text-to-image, llms.txt): Precision and consistent detail for brand assets and product imagery.
    Price: $0.045 per image at default settings (list price $0.05, 10% off)
  3. Qwen Image 3.0 Pro (alibaba/qwen-image-3.0-pro/text-to-image, llms.txt): Advanced prompt understanding for detailed layouts, up to 2K.
    Price: $0.04 per image at default settings
  4. Seedream 5.0 Pro (bytedance/seedream-v5.0-pro, llms.txt): Strong prompt following, 1K or 2K output.
    Price: $0.0405 per image at default settings (list price $0.045, 10% off)

Edit images or combine references

Change part of a photo, restyle it, or merge several reference images into one.

  1. Seedream 5.5 (bytedance/seedream-v5.5/edit, llms.txt): Seedream 5.5 Edit by ByteDance edits and generates images from single-image or multi-reference inputs, supporting up to 10 reference images, consistent multi-turn edits, native transparent-background output, and 1K, 1.5K and 2K output tiers.
    Price: $0.0266 per image with one reference image (list price $0.028, 5% off)
  2. Seedream 5 Pro (bytedance/seedream-v5.0-pro/edit, llms.txt): Seedream V5.0 Pro Edit by ByteDance edits and generates images from single-image or multi-reference inputs, supporting up to 10 reference images, aspect ratio selection, and 1K / 2K output tiers.
    Price: $0.0405 per image with one reference image (list price $0.045, 10% off)
  3. GPT Image 2.6 (openai/gpt-image-2.6-flare/edit, llms.txt): GPT Image 2.6 Flare Edit transforms one or more reference images using natural-language instructions, with a focus on fast editing and rapid creative iteration for visual refinements, marketing assets, product imagery, and production workflows.
    Price: $0.0585 per image with one reference image (list price $0.065, 10% off)
  4. Nano Banana 2.1 (google/nano-banana-2.1/edit, llms.txt): Google Nano Banana 2.1 Edit is the latest Nano Banana image editing model, with gains in visual design, subject consistency and prompt adherence, up to 14 reference images and 1K to 4K output.
    Price: $0.0378 per image with one reference image (list price $0.042, 10% off)
  5. Nano Banana Pro (google/nano-banana-pro/edit, llms.txt): Google Nano Banana Pro (Gemini 3.0 Pro Image) Edit enables image editing with 4K-capable output.
    Price: $0.126 per image with one reference image (list price $0.14, 10% off)

Fast, high-volume generation

Batches, previews and product feeds where speed and unit cost matter most.

  1. GPT Image 2.6 Flare (openai/gpt-image-2.6-flare/text-to-image, llms.txt): Focused on fast generation and rapid creative iteration.
    Price: $0.045 per image at default settings (list price $0.05, 10% off)
  2. Z-Image Turbo (wavespeed-ai/z-image/turbo, llms.txt): Photorealistic images in under a second.
    Price: $0.005 per image at default settings
  3. Qwen Image 3.0 (alibaba/qwen-image-3.0/text-to-image, llms.txt): High-quality images with advanced prompt understanding, up to 2K.
    Price: $0.03 per image at default settings

Upscale and restore images

Make small or soft images sharp enough for print and large screens.

  1. VOSR2 Image Upscaler (wavespeed-ai/vosr2/image, llms.txt): Restores and upscales photos to 2K or 4K in one pass, rebuilding fine structure, faces and small text with faithful color.
    Price: $0.01 per image
  2. Image Upscaler (wavespeed-ai/image-upscaler, llms.txt): Raises resolution to 4K or 8K with more detail and clarity.
    Price: $0.01 per image at default settings

Video

Cinematic text-to-video with sound

Ads, trailers and story clips generated from a prompt, with audio made together with the picture.

  1. Seedance 2.5 (bytedance/seedance-2.5/text-to-video, llms.txt): Seedance 2.5 (Text-to-Video) generates Hollywood-grade cinematic videos from text prompts and optional reference images, videos, and audio with native audio-visual synchronization, director-level camera and lighting control, and exceptional motion stability.
    Price: $0.324 per second of video at 720p (list price $0.36, 10% off)
  2. Seedance 2.5 Flash (bytedance/seedance-2.5-flash/text-to-video, llms.txt): Seedance 2.5 Flash Text-to-Video creates videos from text prompts and optional image, video, and audio references, with 480p to 4K output and optional generated audio.
    Price: $0.216 per second of video at 720p (list price $0.24, 10% off)
  3. MiniMax H3 (wavespeed-ai/minimax-h3/text-to-video, llms.txt): MiniMax H3 Open Weights Text to Video generates coherent videos from text prompts, with 480P / 540P / 768P output, native stereo audio, 5-15 second duration, flexible aspect ratios, and per-second billing.
    Price: $0.04 per second of video at 768p (list price $0.08, 50% off)
  4. WAN 3.0 (alibaba/wan-3.0/text-to-video, llms.txt): Wan 3.0 Text to Video generates cinematic videos from text prompts, with flexible 2-30 second duration, aspect ratio control, optional audio, and deep-thinking controls for high-quality video generation.
    Price: $0.095 per second of video at 720p (list price $0.10, 5% off)
  5. Kling 4.0 (kwaivgi/kling-v4.0/text-to-video, llms.txt): Kling V4.0 Text-to-Video generates cinematic clips of 3 to 30 seconds from a text prompt, with 720p and 1080p output, four aspect ratios including 21:9 ultrawide, and synchronized stereo audio.
    Price: $0.1335 per second of video at 720p (list price $0.157, 15% off)

Animate a still image

Turn a product shot, portrait or keyframe into motion.

  1. Seedance 2.5 (bytedance/seedance-2.5/image-to-video, llms.txt): Seedance 2.5 (Image-to-Video) generates Hollywood-grade cinematic videos from reference images and text prompts with native audio-visual synchronization, director-level camera and lighting control, and exceptional motion stability.
    Price: $0.324 per second of video at 720p (list price $0.36, 10% off)
  2. Seedance 2.5 Flash (bytedance/seedance-2.5-flash/image-to-video, llms.txt): Seedance 2.5 Flash Image-to-Video animates a start image using text instructions, with an optional last frame, 480p to 4K output, and optional generated audio.
    Price: $0.216 per second of video at 720p (list price $0.24, 10% off)
  3. MiniMax H3 (wavespeed-ai/minimax-h3/image-to-video, llms.txt): MiniMax H3 Open Weights Image to Video animates a first-frame image, optionally with last-frame guidance, into coherent 480P / 540P / 768P videos with native stereo audio, 5-15 second duration, flexible aspect ratios, and per-second billing.
    Price: $0.04 per second of video at 768p (list price $0.08, 50% off)
  4. WAN 3.0 (alibaba/wan-3.0/image-to-video, llms.txt): Wan 3.0 Image to Video animates a first-frame image into a cinematic video, with optional last-frame guidance, flexible 2-30 second duration, aspect ratio control, optional audio, and deep-thinking controls for high-quality motion and scene continuity.
    Price: $0.095 per second of video at 720p (list price $0.10, 5% off)
  5. Kling 4.0 (kwaivgi/kling-v4.0/image-to-video, llms.txt): Kling V4.0 Image-to-Video animates a first frame into a clip of 3 to 30 seconds, with an optional last frame and up to 8 in-between keyframes to direct the story, 720p and 1080p output and synchronized stereo audio.
    Price: $0.1335 per second of video at 720p (list price $0.157, 15% off)

Keep the same character or product across shots

Series, brand content and multi-shot stories where the subject must stay the same.

  1. Seedance 2.5 (bytedance/seedance-2.5/reference-to-video, llms.txt): Seedance 2.5 Reference to Video creates cinematic videos with native audio from your own reference images, videos, and audio, keeping characters, products, motion, and sound consistent while a prompt directs the new shot.
    Price: $0.324 per second of video at 720p (list price $0.36, 10% off)
  2. Seedance 2.5 Flash (bytedance/seedance-2.5-flash/reference-to-video, llms.txt): Seedance 2.5 Flash Reference-to-Video creates videos guided by text and optional image, video, and audio references, with 480p to 4K output and optional generated audio.
    Price: $0.216 per second of video at 720p (list price $0.24, 10% off)
  3. MiniMax H3 (wavespeed-ai/minimax-h3/reference-to-video, llms.txt): MiniMax H3 Open Weights Reference to Video generates coherent 480P / 540P / 768P videos from prompts and multimodal references, guided by up to 9 reference images, 3 reference videos, and 3 reference audios, with native stereo audio and flexible…
    Price: $0.05 per second of video at 768p (list price $0.10, 50% off)
  4. WAN 3.0 (alibaba/wan-3.0/reference-to-video, llms.txt): Wan 3.0 Reference to Video creates coherent videos from prompts and multimodal references, including images, videos, and audio, with flexible 2-30 second duration and aspect ratio control for subject consistency, motion guidance, timing control, and scene…
    Price: $0.095 per second of video at 720p (list price $0.10, 5% off)
  5. Kling 4.0 (kwaivgi/kling-v4.0/reference-to-video, llms.txt): Kling V4.0 Reference-to-Video builds a new clip of 3 to 30 seconds from up to 10 reference images, 5 reference videos and 3 reference audios, following the subjects, motion and camera you point to in the prompt, with 720p and 1080p output and synchronized…
    Price: $0.1335 per second of video at 720p (list price $0.157, 15% off)

Fast, low-cost drafts

Iterate on a prompt or storyboard before spending on the final render.

  1. Seedance 2.5 Flash Turbo (bytedance/seedance-2.5-flash/text-to-video-turbo, llms.txt): Seedance 2.5 Flash in its turbo tier, 720p or 1080p with optional generated audio.
    Price: $0.126 per second of video at 720p (list price $0.14, 10% off)
  2. Seedance 2.5 Turbo (bytedance/seedance-2.5/text-to-video-turbo, llms.txt): A faster, more affordable tier of Seedance 2.5 at 720p or 1080p.
    Price: $0.18 per second of video at 720p (list price $0.20, 10% off)
  3. MiniMax H3 Open Weights (wavespeed-ai/minimax-h3/text-to-video, llms.txt): Video with native stereo audio, 5 to 15 seconds, billed per second.
    Price: $0.04 per second of video at 768p (list price $0.08, 50% off)
  4. Kling 4.0 Flash (kwaivgi/kling-v4.0-flash/image-to-video, llms.txt): Image-to-video with synchronized audio, focused on fast generation and rapid iteration.
    Price: $0.094 per second of video at 720p

Talking avatars and lip-sync

Presenters, explainers, dubbing and singing portraits driven by an audio track.

  1. InfiniteTalk (wavespeed-ai/infinitetalk, llms.txt): InfiniteTalk Image-to-Video converts a single photo and audio track into long-form audio-driven talking or singing avatar videos, supporting up to 10 minutes, 720P output, natural lip sync, expressive motion, and digital human video workflows.
    Price: Billed per second of the audio you upload.
  2. Seedance 2.5 (bytedance/seedance-2.5/talking-avatar, llms.txt): Seedance 2.5 Talking Avatar animates a portrait image from an audio recording to generate natural talking-avatar videos in 480P or 720P, processing up to the first 120 seconds of audio for digital human, presenter, dubbing, and avatar workflows.
    Price: Billed per second of the audio you upload.
  3. Seedance 2.5 Flash (bytedance/seedance-2.5-flash/talking-avatar, llms.txt): Seedance 2.5 Flash Talking Avatar animates a portrait image from an audio recording to generate natural talking-avatar videos in 480p or 720p, processing up to the first 120 seconds of audio for digital human, presenter, dubbing, and avatar workflows.
    Price: Billed per second of the audio you upload.
  4. LongCat Avatar 1.6 (wavespeed-ai/longcat-avatar-1.6, llms.txt): LongCat Avatar 1.6 turns a portrait image and audio into talking or singing avatar videos, with natural facial expressions and consistent character appearance for digital humans, virtual presenters, social content, and avatar video workflows.
    Price: Billed per second of the audio you upload.

Edit or extend a video you have

Change what is in a clip, or continue it.

  1. Seedance 2.5 Video Edit (bytedance/seedance-2.5/video-edit, llms.txt): Rewrites lighting, style, weather, environment or specific elements while the source video keeps identity, composition and motion.
    Price: Billed per second of the video you upload.
  2. Kling 4.0 Video Edit (kwaivgi/kling-v4.0/video-edit, llms.txt): Rewrites a 3 to 30 second video from a text instruction, keeping the scene, subject and motion.
    Price: Billed per second of the video you upload.
  3. Seedance 2.5 Video Extend (bytedance/seedance-2.5/video-extend, llms.txt): Continues a video from its last frame with a new cinematic shot from a prompt.
    Price: $0.198 per second at 720p, counting both the input video (at least 2 seconds) and the new seconds: extending a 10-second clip by 5 seconds bills 15 seconds (list price $0.22, 10% off)

Upscale and restore videos

Make low-resolution or AI-generated clips sharp enough for 4K delivery.

  1. Ultimate Video Upscaler (wavespeed-ai/ultimate-video-upscaler, llms.txt): Our highest-quality video upscaler: 720p to 4K with stable, flicker-free frames and rebuilt fine detail such as hair, fabric and foliage, well suited to AI-generated clips.
    Price: Billed per second of the video you upload, at a rate set by the output resolution.
  2. Video Upscaler Pro (wavespeed-ai/video-upscaler-pro, llms.txt): Upscales video to 720p, 1080p, 2K or 4K with seamless motion and frame-to-frame consistency.
    Price: Billed per second of the video you upload, at a rate set by the output resolution.

Copy motion or swap characters

Dances, gestures and actions applied to your own character, or a character swapped into an existing video.

  1. Seedance 2.5 Motion Control (bytedance/seedance-2.5/motion-control, llms.txt): Your character, from a picture or a description, performs the movements, gestures and camera motion of a reference video, up to 1080p.
    Price: Billed per second of the reference video.
  2. Seedance 2.5 Flash Motion Control (bytedance/seedance-2.5-flash/motion-control, llms.txt): Transfers the movements, gestures and camera motion of a reference video onto a character from a picture or a description, up to 1080p.
    Price: Billed per second of the reference video.
  3. Wan 2.2 Animate 2 (wavespeed-ai/wan-2.2/animate-2, llms.txt): Makes the character in a reference image perform the motion of a driving video, with no pose extraction.
    Price: Billed per second of the driving video.
  4. Openjutsu Character Swap (wavespeed-ai/openjutsu/character-swap, llms.txt): Replaces people in an existing video with characters from reference images, keeping the scene, motion, framing and audio.
    Price: Billed per second of the video you upload, plus a fixed charge per reference image.

4K video

Final deliverables at the highest resolution.

  1. Seedance 2.5 (bytedance/seedance-2.5/text-to-video, llms.txt): Cinematic video up to 4K with native audio-visual sync.
    Price: $1.62 per second of video at 4K (list price $1.80, 10% off)
  2. Seedance 2.5 Flash (bytedance/seedance-2.5-flash/text-to-video, llms.txt): Text-to-video from 480p to 4K with optional generated audio.
    Price: $1.08 per second of video at 4K (list price $1.20, 10% off)
  3. Kling O3 4K (kwaivgi/kling-video-o3-4k/text-to-video, llms.txt): Cinematic 4K video from text with subject consistency.
    Price: $0.42 per second of 4K video

Audio

Text-to-speech and voice cloning

Narration, dialogue, dubbing and voice agents.

  1. ElevenLabs v4 (elevenlabs/eleven-v4, llms.txt): Expressive multilingual speech with voice, stability and similarity controls.
    Price: $0.08 per 1,000 characters
  2. MiniMax Speech 2.6 HD (minimax/speech-2.6-hd, llms.txt): Low-latency speech with voice cloning in 40+ languages.
    Price: $0.10 per 1,000 characters
  3. OmniVoice (wavespeed-ai/omnivoice/text-to-speech, llms.txt): Zero-shot speech in 600+ languages.
    Price: $0.05 per 1,000 characters
  4. Qwen3 TTS Voice Clone (wavespeed-ai/qwen3-tts/voice-clone, llms.txt): Clones a voice from reference audio and speaks new text with it.
    Price: $0.05 per 1,000 characters

Music and songs

Background tracks, jingles and full songs with vocals.

  1. Lyria 3.5 (google/lyria-3.5/music, llms.txt): Full songs of up to about three minutes with expressive vocals and timed lyrics.
    Price: $0.10 per run at default settings
  2. ElevenLabs Music v2.5 (elevenlabs/music-v2.5, llms.txt): Songs with vocals or instrumental music from a prompt.
    Price: $0.75 per run at default settings

3D

Image to 3D model

Game assets, product models and prototypes from a picture.

  1. Meshy 7.1 (meshy/v7.1/image-to-3d, llms.txt): Meshy 7.1 Image-to-3D turns a single image into a textured 3D model.
    Price: $1.20 per model at default settings
  2. Tripo3D H3.1 (tripo3d/h3.1/image-to-3d, llms.txt): Tripo3D H3.1 Image-to-3D turns a single image into a 3D model with Tripo's high-fidelity H3.1 model (up to 2 million faces) for production assets and 3D printing.
    Price: $0.36 per model at default settings
  3. Tripo3D P2 (tripo3d/p2/image-to-3d, llms.txt): Tripo3D P2 Image-to-3D turns a single image into a 3D model with Tripo's P2 model with native quad topology (up to 50,000 triangles or 25,000 quads) for games and real-time pipelines.
    Price: $1.32 per model at default settings

Text to 3D model

Props, characters and concepts described in words.

  1. Meshy 7.1 (meshy/v7.1/text-to-3d, llms.txt): Meshy 7.1 Text-to-3D turns a text description into a textured 3D model.
    Price: $1.20 per model at default settings
  2. Tripo3D H3.1 (tripo3d/h3.1/text-to-3d, llms.txt): Tripo3D H3.1 Text-to-3D turns a text description into a 3D model with Tripo's high-fidelity H3.1 model (up to 2 million faces) for production assets and 3D printing.
    Price: $0.24 per model at default settings
  3. Tripo3D P2 (tripo3d/p2/text-to-3d, llms.txt): Tripo3D P2 Text-to-3D turns a text description into a 3D model with Tripo's P2 model with native quad topology (up to 50,000 triangles or 25,000 quads) for games and real-time pipelines.
    Price: $1.32 per model at default settings

Rig, animate and clean up 3D models

Make a generated model ready for a game engine or an animation.

  1. Meshy Animation (meshy/animation, llms.txt): Rigs a character model and applies up to 10 animations from the Meshy library.
    Price: $0.32 per run
  2. Tripo3D Auto Rig (tripo3d/rig, llms.txt): Adds a skeleton to a humanoid, animal or creature model, ready for animation.
    Price: $0.30 per run
  3. Meshy Rigging (meshy/rigging, llms.txt): Adds a skeleton to a humanoid or four-legged character model.
    Price: $0.20 per run
  4. Tripo3D Retopology (tripo3d/retopology, llms.txt): Rebuilds a model with fewer, cleaner polygons, as triangles or quads.
    Price: $0.42 per run
  5. Meshy Remesh (meshy/remesh, llms.txt): Rebuilds a model at a target polycount, as triangles or quads.
    Price: $0.20 per run

Language models (LLM API)

137 language models (GPT, Claude, Gemini, DeepSeek, Qwen, Grok, Kimi, GLM and more) through one OpenAI-compatible endpoint. Billed per token at the published rates; the same API key as the media models.

  • Endpoint: https://llm.wavespeed.ai/v1. With an OpenAI SDK, set base_url to it and the API key to your WaveSpeed key; call chat.completions.create with a model id below.
  • Every model, its context length and prices: wavespeed.ai/llm. Try them side by side: LLM Playground.

Flagship models

Live prices per 1M tokens at the base tier; some models charge more for long context or less for cached input (see each model page).

  • Claude Opus 5.5 (anthropic/claude-opus-5.5, 1M context): $4.00 input / $20.00 output per 1M tokens
  • Claude Sonnet 5.5 (anthropic/claude-sonnet-5.5, 1M context): $2.00 input / $10.00 output per 1M tokens
  • Claude Fable 5.2 (anthropic/claude-fable-5.2, 1M context): $8.00 input / $40.00 output per 1M tokens
  • GPT-6 Astra (openai/gpt-6-astra, 1.05M context): $10.00 input / $50.00 output per 1M tokens
  • GPT-6.1 Sol (openai/gpt-6.1-sol, 1.05M context): $2.00 input / $10.00 output per 1M tokens
  • GPT-6 Luna (openai/gpt-6-luna, 1.05M context): $0.10 input / $0.50 output per 1M tokens
  • Gemini 4 Carbon (google/gemini-4-carbon, 1.05M context): $4.00 input / $20.00 output per 1M tokens
  • Grok 4.7 (x-ai/grok-4.7, 500K context): $2.00 input / $6.00 output per 1M tokens
  • Kimi K3.1 Pro (moonshotai/kimi-k3.1-pro, 1.05M context): $2.00 input / $10.00 output per 1M tokens
  • DeepSeek V4.1 Flash 1017 (deepseek/deepseek-v4.1-flash-1017, 1.05M context): $0.15 input / $0.60 output per 1M tokens

All models by type

Every public image, video, audio and 3D model, grouped by what it takes and returns (the language models are on /llm). Each links to its page (inputs, price, playground); add /llms.txt to that URL for its markdown summary (description, inputs, price, examples).

Image to Video (186)

Text to Image (161)

Text to Video (128)

Image to Image (126)

Video to Video (59)

Image to 3D (53)

Digital Human (49)

Reference to Video (47)

Text to Speech (38)

AI Music (29)

Image Tools (26)

Video Extend (25)

Video Tools (20)

Upscale Image (18)

Motion Control (17)

Upscale Video (16)

Image Effects (14)

Text to 3D (13)

Training (13)

Image to Text (11)

Video to Audio (10)

Sound Effects (8)

Image Background Removal (7)

Video Effects (7)

Speech to Text (6)

Content Moderation (5)

Image Face Swap (5)

Image Object Removal (5)

LLM (5)

Video to Text (5)

Audio Tools (4)

Video Object Removal (4)

Lyrics Generation (3)

Music Analysis (3)

Video Background Removal (3)

Video Face Swap (3)

Voice Cloning (3)

Music Editing (2)

Speech to Speech (2)

Video Dubbing (2)

Audio to Audio (1)

Audio to Video (1)