Seedance 2.0
Integrate the Seedance 2.0 text-to-video model to render hyper-realistic action scenes using our developer-friendly AI API.
curl --request POST "https://gptproto.com/api/v3/doubao/doubao-seedance-2-0-fast-260128/text-to-video" \
--header "Authorization: Bearer $GPTPROTO_API_KEY" \
--header "Content-Type: application/json" \
--data '{
"prompt": "A tiny origami fox sailing a teacup across a moonlit puddle",
"aspect_ratio": "16:9",
"duration": 5,
"resolution": "720p",
"generate_audio": true,
"camera_fixed": false,
"seed": -1
}'Start from the cost of a single sample and pick a testing budget.
$0.2219 / image
If you're looking to push the boundaries of motion in digital content, you can browse Seedance 2.0 and other models on our platform to see why this ByteDance-backed tool is causing such a stir. Seedance 2.0 Fast isn't just another video generator; it's a specialized engine designed for high-intensity action and realism that many other models struggle to replicate.
Seedance 2.0 Fast is the low-latency tier of ByteDance's Seedance 2.0 video family, tuned to return high-action text-to-video clips faster than the full 2.0 model while keeping the same motion engine. On GPTProto the model is served as doubao-seedance-2-0-fast-260128 and takes a text prompt plus a small set of controls: aspect ratio, duration (4–14s), resolution (480p / 720p), a native audio toggle, a camera-fixed option, and a seed for reproducibility. It is a text-to-video endpoint — you describe the shot, and the Seedance 2.0 Fast API renders the motion.
| Spec | Value |
|---|---|
| Model string | doubao-seedance-2-0-fast-260128 |
| Modality | Text-to-video |
| Resolution | 480p / 720p |
| Duration | 4–14 s |
| Aspect ratio | 16:9 (+ presets) |
| Native audio | Yes (toggle) |
| Camera control | Camera-fixed option |
| Reproducibility | Seed |
| GPTProto price | From $0.2365 / clip |
| Underlying model | ByteDance Seedance 2.0 (Doubao channel) |
You'll see two Seedance 2.0 Fast listings on GPTProto: doubao-seedance-2-0-fast-260128 and dreamina-seedance-2-0-fast-260128. They run the same Seedance 2.0 Fast weights — the difference is the upstream ByteDance channel each one routes to. The doubao build goes through ByteDance's Doubao / Volcano Engine (Ark) developer platform; the dreamina build is the Dreamina creative-line version, which ByteDance exposes internationally through BytePlus ModelArk. Both list from $0.2365 on GPTProto and share the same key. For most text-to-video work they're interchangeable; we list both so you can switch channels if one is throttled or unavailable. See the dreamina-seedance-2-0-fast-260128 page for the full Dreamina-vs-Doubao decision guide.
| doubao-…-fast-260128 (this page) | dreamina-…-fast-260128 | |
|---|---|---|
| Model weights | Seedance 2.0 Fast (same) | Seedance 2.0 Fast (same) |
| Upstream channel | Doubao / Volcano Engine (Ark) | Dreamina line · BytePlus ModelArk (intl) |
| GPTProto price | $0.2365 from | $0.2365 from |
| API key | Same key, 200+ models | Same key |
| Practical difference | Channel availability / routing | Channel availability / routing |
Not sure which Seedance tier to call? All of these are live on GPTProto under one key, so the choice is about output tier, not access. Seedance 2.0 Fast and Seedance 2.0 Mini share the same list price; Mini is the lighter-weight variant aimed at quick drafts, while Fast prioritizes high-action motion. The full Seedance 2.0 costs more per clip for maximum fidelity, and Seedance 1.5 Pro is the budget option from a previous generation.
| Model (GPTProto string) | Tier | From price | Use it for |
|---|---|---|---|
doubao-seedance-2-0-fast-260128 |
Fast (this page) | $0.2365 | Fast high-action t2v |
dreamina-seedance-2-0-mini-260615 |
Mini | $0.2365 | Quick low-cost drafts |
doubao-seedance-2-0-260128 |
Full 2.0 | $0.2957 | Max fidelity |
seedance-1-5-pro-251215 |
1.5 Pro | $0.0408 | Budget / prior gen |
Seedance 2.0 Fast is billed per clip on GPTProto, so pricing scales with resolution and duration rather than an opaque monthly credit bundle. A 480p clip starts at $0.2475 for 4 seconds and rises about $0.06 for each additional second, so you always know a render's cost before you send it. To keep spend affordable on heavy runs, draft first: generate a 4–6 second 480p preview to lock your prompt, motion, and framing, then commit only the approved shot to a longer 720p render. Because the same API key spans 200+ models, there's no separate top-up or credit maze per model.
Where Seedance 2.0 Fast stands out is motion consistency. In high-intensity shots — a parkour run, a car chase, a fast camera push — movement stays weighted and physically plausible instead of the floaty, disjointed drift common in other text-to-video models. That's the same Seedance 2.0 motion engine that powers ByteDance's Doubao and Dreamina products. It's a genuine trade-off, not magic: Fast favors dynamic action over frame-perfect likeness, which is exactly why it suits trailers, sports, and social action content.
Two things worth knowing before you scale up. First, prompts: Seedance 2.0 Fast responds best to concrete motion and camera language — name the action, the speed, and the shot ("low-angle tracking shot, sprinter accelerating, motion blur") rather than static scene description. Set duration to match the beat you want and use the seed to iterate on a shot you like. Second, grain: some outputs carry a fine film-like grain. Many creators keep it for a cinematic look; if you need a clean plate, run the clip through an AI upscaler as a post step. Neither is a blocker — they're knobs, and knowing them saves failed renders.
Reaching the Seedance 2.0 Fast API directly through ByteDance's channels is hard for developers outside China — it often needs a Chinese phone number and a local payment method. GPTProto is the bridge: sign up with email, and you get doubao-seedance-2-0-fast-260128 access behind one OpenAI-compatible API key that also covers 200+ other image, video, and text models on the same balance. That means English docs, one integration, transparent per-clip pricing, and no regional gateway to fight — so you can ship the Seedance 2.0 Fast API into your product today rather than waiting on account approval.
Guides, comparisons, and updates related to this model.
All Articles
Use 17 copy-ready Seedream 5.0 Pro prompts for realistic photos, controlled edits, product images, posters, and API workflows—without unwanted identity or layout drift.

A developer's guide to Kling 3.0 Motion Control — pro vs std, input limits, prompt tips, and runnable API code (Python + cURL) via GPTProto.

Seedance 2.0 Mini isn't "half price"—on the API it's ~20% cheaper. The big saving comes from 720p drafting. Real pricing, runnable code, which tier to ship.

Nano Banana 2 costs half of Nano Banana Pro and scores higher on the Image Arena. See when each Gemini image model wins—plus one-API code to run both.
Input
Output