Generated from packages/schema/src/v2/models.ts. Call the list_models MCP tool (or GET /v1/models) for the live version.
Default to seedance-2.0. It is right for talking-head UGC, product-in-hand, animating a still and crazy-look at 30 credits/s at 480p, 60 at 720p, 150 at 1080p. seedance-2.5: about 0.9x the credits at 480p (28 credits/s at 480p, 60 at 720p, 150 at 1080p); pick it only when the user asks for the best possible single clip and accepts the wait. Never for drafts or bulk. Default quality: 480p on seedance-2.5, 720p on seedance-2.0; 480p is the cheap draft, 1080p the dear finish.
Pass the id as model to generate_video / generate_image / generate_audio (MCP) or POST /v2/generate/<kind> (REST). Omit it for the default, or pass "auto". Live video models are also the engine of the fixed REST skills (/v2/selfie, /v2/crazy-look, CLI --engine). Only live models are accepted; a planned id returns a 400 naming the live ones.
The mode follows from the fields: first_frame (and optional last_frame) is image-to-video, refs / video_refs / audio_refs is reference, neither is text. Frames and refs cannot be mixed on Seedance. Reference clip seconds (video_refs) are billed at the same per-second rate as output seconds. Credits below are per output second.
| Model | Mode | Inputs | Seconds | Aspects | Credits/s | Verified |
|---|---|---|---|---|---|---|
| seedance-2.0 | text | prompt only | 4 to 15 | 9:16, 16:9, 1:1, 4:3, 3:4, 21:9, adaptive (default 9:16) | 30 at 480p, 60 at 720p, 150 at 1080p | 2026-09-06: 4s, 16:9, 480p, native audio; rendered in 231 s |
| seedance-2.0 | image | first_frame + optional last_frame | 4 to 15 | 9:16, 16:9, 1:1, 4:3, 3:4, 21:9, adaptive (default adaptive) | 30 at 480p, 60 at 720p, 150 at 1080p | 2026-09-06: first_frame only (4s, 480p, adaptive) and first_frame + last_frame (task-unified-1788681109-w09254uu); both rendered in about 3 to 5 min |
| seedance-2.0 | reference | refs up to 9, video_refs up to 3 (15 s total), audio_refs up to 3 (15 s total); audio needs an image or video beside it | 4 to 15 | 9:16, 16:9, 1:1, 4:3, 3:4, 21:9, adaptive (default 9:16) | 30 at 480p, 60 at 720p, 150 at 1080p | 2026-09-06: image ref (4s, 480p) and image ref + 8 s reference clip (task-unified-1788681110-hdbufncn); earlier 5s 720p portrait run b0011e92 via Claude Code |
| seedance-2.5 | text | prompt only | 4 to 30 | 9:16, 16:9, 1:1, 4:3, 3:4, 21:9, adaptive (default 9:16) | 28 at 480p, 60 at 720p, 150 at 1080p | 2026-09-05: generate_video, text only, 4s, 720p, 396 credits; the provider took about 25 min |
| seedance-2.5 | image | first_frame + optional last_frame | 4 to 30 | adaptive (default adaptive) | 28 at 480p, 60 at 720p, 150 at 1080p | 2026-09-06: first_frame, 4s, 480p, adaptive; rendered in 262 s |
| seedance-2.5 | reference | refs up to 30, video_refs up to 10 (30 s total), audio_refs up to 10 (30 s total) | 4 to 30 | 9:16, 16:9, 1:1, 4:3, 3:4, 21:9, adaptive (default 9:16) | 28 at 480p, 60 at 720p, 150 at 1080p | 2026-09-06: generate_video with a product image ref via Claude Code, 8s, 720p (perfume UGC run) |
Notes for seedance-2.0:
- text: No seed. Every render is new; keep a series consistent with references, not seeds.
- image: first_frame becomes frame one of the clip; last_frame (optional) becomes the final frame and the model animates between them.
- image: Frames only: refs, video_refs and audio_refs are not accepted in this mode. To keep an identity AND set the frame, put the frame image in refs and describe it as @image1.
- image: aspect "adaptive" (default) follows the first frame. Frame images: jpeg/png/webp, ratio between 0.4 and 2.5, 300 to 6000 px, up to 30 MB.
- reference: Address references in the prompt: "@image1 holds @image2 and says ...". Numbering is per list: refs are @image1.., video_refs are @video1.., audio_refs are @audio1...
- reference: Reference clips: 2 to 15 s each, 15 s in total, mp4/mov 480p to 1080p, 24 to 60 fps, up to 50 MB. Their seconds are billed like output seconds.
- reference: Audio references: wav/mp3, 2 to 15 s each, 15 s in total, up to 15 MB. Audio alone is not accepted: add an image or video reference.
- reference: aspect "adaptive" follows the first video reference, else the first image, else the prompt.
Notes for seedance-2.5:
- text: No seed.
- image: first_frame becomes frame one; last_frame (optional) becomes the final frame.
- image: aspect must be "adaptive" on this model in image mode: the clip takes the frame's ratio. Any other aspect is refused.
- image: Frames only: refs, video_refs and audio_refs are not accepted in this mode.
- reference: Address references as @image1.., @video1.., @audio1.. (numbered per list).
- reference: Reference clips: 2 to 30 s each, 30 s in total, up to 4K, up to 200 MB; their seconds are billed like output seconds. Audio references may stand alone here.
- reference: Never write edit or extend intent ("edit the video", "add", "remove", "replace", "extend", "continue") into a reference prompt: the provider reclassifies the task and fails it minutes later. Describe the NEW clip you want.
list_models / GET /v1/models add a recent block per model: the last 30 days of loose-surface runs, runs, fail_rate, auto_score (0 to 1, an auto-judge grades every job: 3 frames or the image against the realism rubric, prompt adherence, identity match when refs were given), scored, user_score (1 to 5 from rate_run), rated, p50_seconds. null until a model has run.
model: "auto" is one printed rule over those numbers: the kind's default, unless it failed more than 25% of at least 10 recent runs and another live model is healthy, or a live model within 1.5x the default's price beats its auto score by 0.10 or more over at least 10 judged runs. Only models that have the request's mode are candidates. The quote and the submit response say which model was chosen and why.
| Model | Kind | Tier | Price | Modes and limits | Best for | Avoid for |
|---|---|---|---|---|---|---|
| seedance-2.0 (default) | video | standard | 30 credits/s at 480p, 60 at 720p, 150 at 1080p | text (prompt only): 4 to 15 s, default aspect 9:16, 480p/720p/1080p; image-to-video (first_frame + optional last_frame): 4 to 15 s, default aspect adaptive, 480p/720p/1080p; reference (refs / video_refs / audio_refs, up to 9 images, 3 clips, 15 s total, 3 audio, 15 s total, not alone): 4 to 15 s, default aspect 9:16, 480p/720p/1080p | talking-head UGC; product in hands; crazy look; bulk daily posts; animating a still (first frame) into a clip | clips over 15s; hero shots where 2.5 detail is worth the longer wait; anything that needs a seed |
| seedance-2.5 | video | premium | 28 credits/s at 480p, 60 at 720p, 150 at 1080p | text (prompt only): 4 to 30 s, default aspect 9:16, 480p/720p/1080p; image-to-video (first_frame + optional last_frame): 4 to 30 s, aspect adaptive only, 480p/720p/1080p; reference (refs / video_refs / audio_refs, up to 30 images, 10 clips, 30 s total, 10 audio, 30 s total): 4 to 30 s, default aspect 9:16, 480p/720p/1080p | hero product ads; close-up faces; one clip that has to be the best | drafts; bulk; anyone who cannot wait 15 to 30 minutes |
| gpt-image-2.5 (default) | image | premium | 20 credits per image | 1024x1024, 1024x1536, 1536x1024; refs up to 4 | portraits and character sheets that a video has to keep; the first frame of a clip; product in hand; edits that must not lose the face | bulk throwaway drafts where gpt-image-2.5-flare is faster |
| gpt-image-2.5-flare | image | standard | 20 credits per image | 1024x1024, 1024x1536, 1536x1024; refs up to 4 | variants and drafts at the same quality tier; batches of frames; anything where a few seconds matter | the one sheet a whole series depends on, where gpt-image-2.5 edits hold identity a little better |
| gpt-image-2 | image | standard | 20 credits per image | 1024x1024, 1024x1536, 1536x1024; refs up to 4 | the previous generation, kept selectable for runs that were built on it | new work: gpt-image-2.5 is the default and holds identity better |
| elevenlabs-tts (default) | audio | standard | 1 credit per 100 characters | up to 4000 characters per call | voiceover on b-roll; narration; a standalone voice file; an audio reference for generate_video | lip-synced talking head: generate_video renders speech natively |
From the catalog usage card. list_models returns the same text plus the last 30 days of real results.
- Pick it when you need a real-looking person saying real words, or a still brought to life, at a normal budget.
- Best for: talking-head UGC; product in hands; crazy look; bulk daily posts; animating a still (first frame) into a clip.
- Avoid for: clips over 15s; hero shots where 2.5 detail is worth the longer wait; anything that needs a seed.
- Latency: about 3 minutes for a 5 s clip at 720p.
- Price: 30 credits/s at 480p, 60 at 720p, 150 at 1080p; reference clip seconds (video_refs) are billed at the same per-second rate as output seconds.
- Modes:
- text (prompt only): 4 to 15 s, default aspect 9:16, 480p/720p/1080p
- image-to-video (first_frame + optional last_frame): 4 to 15 s, default aspect adaptive, 480p/720p/1080p
- reference (refs / video_refs / audio_refs, up to 9 images, 3 clips, 15 s total, 3 audio, 15 s total, not alone): 4 to 15 s, default aspect 9:16, 480p/720p/1080p
- Prompting:
- Write the shot as a director: who (age, look), where (setting, light), what happens, phone framing, and the exact spoken words in quotes. About 2.3 words per second.
- Keep a person consistent across clips with refs (a portrait or a character sheet), addressed as @image1; the model has no seed.
- To animate a specific still, pass it as first_frame; add last_frame to control where the motion ends.
- Pick it when the user asked for the best possible single clip and accepts the wait (12 to 25 minutes); at 480p it costs the same as 2.0.
- Best for: hero product ads; close-up faces; one clip that has to be the best.
- Avoid for: drafts; bulk; anyone who cannot wait 15 to 30 minutes.
- Latency: 12 to 25 minutes per clip; plan the wait.
- Price: 28 credits/s at 480p, 60 at 720p, 150 at 1080p; reference clip seconds (video_refs) are billed at the same per-second rate as output seconds.
- Modes:
- text (prompt only): 4 to 30 s, default aspect 9:16, 480p/720p/1080p
- image-to-video (first_frame + optional last_frame): 4 to 30 s, aspect adaptive only, 480p/720p/1080p
- reference (refs / video_refs / audio_refs, up to 30 images, 10 clips, 30 s total, 10 audio, 30 s total): 4 to 30 s, default aspect 9:16, 480p/720p/1080p
- Prompting:
- Same director-style prompt and @image1 references as seedance-2.0.
- In image mode leave aspect out (it is adaptive); the frame decides the ratio.
- Do not put edit or extend wording in a reference prompt.
- Pick it when you are making the image a video will be built on: a portrait, a character sheet, a first frame, or an edit that has to keep the same person.
- Best for: portraits and character sheets that a video has to keep; the first frame of a clip; product in hand; edits that must not lose the face.
- Avoid for: bulk throwaway drafts where gpt-image-2.5-flare is faster.
- Latency: about 30 seconds, a little longer for an edit with references.
- Price: 20 credits per image.
- Prompting:
- Concrete subject, age, framing, light, what the hands do; it follows layout instructions like "four poses on a plain background" or "headroom for a caption".
- With refs it edits or composes from them and holds the identity across poses, which is what makes a character sheet usable as a video reference.
- Pass the result straight to generate_video as first_frame (to animate it) or in refs (to keep that person across clips).
- Pick it when you want the same look as gpt-image-2.5 but faster, or you are making several images at once.
- Best for: variants and drafts at the same quality tier; batches of frames; anything where a few seconds matter.
- Avoid for: the one sheet a whole series depends on, where gpt-image-2.5 edits hold identity a little better.
- Latency: about 15 to 30 seconds.
- Price: 20 credits per image.
- Prompting:
- Same prompts as gpt-image-2.5; it is the speed tier of the same family.
- Pick it when you are reproducing something that was made on gpt-image-2; otherwise take the default.
- Best for: the previous generation, kept selectable for runs that were built on it.
- Avoid for: new work: gpt-image-2.5 is the default and holds identity better.
- Latency: under a minute.
- Price: 20 credits per image.
- Prompting:
- Concrete subject, age, framing, light, what the hands do; it follows layout instructions like "headroom for a caption".
- With refs it EDITS or composes from them (a product into a hand, a portrait re-lit); without refs it paints from the prompt alone.
- Pick it when you need a clean voice track and no face.
- Best for: voiceover on b-roll; narration; a standalone voice file; an audio reference for generate_video.
- Avoid for: lip-synced talking head: generate_video renders speech natively.
- Latency: seconds.
- Price: 1 credit per 100 characters.
- Prompting:
- Emotion tags like [excited] or [whispers] are honoured; keep sentences short for pacing.
- Seven named voices (sarah default) or a raw ElevenLabs voice id.
Each goes live only after a real run is recorded and a user price is set.
| Model | Kind | Tier | Best for |
|---|---|---|---|
| seedance-2.0-mini | video | draft | drafts |
| kling-o3 | video | premium | 1080p hero shots with a non-Seedance look |
| wan-3.0 | video | standard | single takes of 15 to 30 s |
| omnihuman-1.5 | video | premium | lip-syncing one photo to an existing recording |
| sora-2 | video | premium | cinematic b-roll without a locked face |
| nano-banana-2 | image | standard | product placement into a character frame |
| seedream-5.0-pro | image | premium | multi-reference composites: person + product + setting |
| z-image-turbo | image | draft | framing wireframes |
| doubao-seed-audio-1.0 | audio | draft | cloning a voice from a short reference clip |
| suno | audio | standard | a music bed under a clip |
100 credits = 1 USD.