Seedance 2.0 Mini (dreamina-seedance-2-0-mini-260615) is ByteDance's budget-tier text-to-video model, derived from the Seedance 2.0 architecture. Where the full Seedance 2.0 model targets maximum fidelity and multi-reference control, Mini trades part of that ceiling for lower per-run cost and faster turnaround — it's the variant you reach for when you need many clips, not one polished final. From a text prompt it produces 4–14 second clips at 480p or 720p with native audio and a camera-lock toggle, making it a practical Seedance 2.0 Mini video generator for high-frequency social and ad workflows.
| Spec | Seedance 2.0 Mini |
|---|---|
| Provider | ByteDance (Dreamina) |
| Model string | dreamina-seedance-2-0-mini-260615 |
| Task | Text-to-video |
| Input | Text prompt |
| Output | MP4 video, optional native audio |
| Resolution | 480p, 720p |
| Duration | 4–14s (same range at 480p and 720p) |
| Aspect ratios | 16:9, 9:16, 4:3, 3:4, 1:1 and other standard ratios |
| Native audio | Yes (generate_audio) |
| Camera control | camera_fixed toggle + prompt-driven moves |
| Seed | Supported (seed, -1 = random) |
| GPTProto price | from $0.2365/run (varies by duration & resolution) |
| Endpoint | POST /api/v3/doubao/dreamina-seedance-2-0-mini-260615/text-to-video |
Need image input or first/last-frame control? Image-to-video runs as a separate task on its own page: Seedance 2.0 Mini image-to-video. This page covers text-to-video.
Seedance 2.0 Mini vs Seedance 2.0, Seedance 1.5 Pro & Veo 3
All four models below are live on GPTProto under one balance, so you switch between them by changing the model string — no new account, no separate billing.
| Seedance 2.0 Mini | Seedance 2.0 | Seedance 1.5 Pro | Veo 3 | |
|---|---|---|---|---|
| Model string | dreamina-seedance-2-0-mini-260615 |
dreamina-seedance-2-0-260128 |
seedance-1-5-pro-251215 |
veo3 |
| Generation | Text-to-video | Text / image / audio / video → video | Text- & image-to-video | Text- & image-to-video |
| Max duration | 14s | 15s | 12s | 8s |
| Resolution | 480p / 720p | up to 4K | 720p | 720p / 1080p |
| Native audio | Yes | Yes | Yes | Yes |
| GPTProto price | from $0.2365/run | from $0.2957/run | from $0.0408/run | $0.48/run |
| Best for | High-volume text-to-video | 4K multi-reference hero shots | Lowest-cost workhorse | Audio-led 8s shots |
Seedance 2.0 Mini vs Seedance 2.0 (260128)
Same generation family. The full 260128 model pushes the ceiling to 4K resolution, a 15-second max, and richer multi-reference inputs — for about 25% more per run ($0.2957 vs $0.2365). Mini stops at 720p and 14s. Use Mini for volume and iteration; step up to Seedance 2.0 when the take needs 4K or has to be the hero shot.
Seedance 2.0 Mini vs Seedance 1.5 Pro
This one isn't about cost — Seedance 1.5 Pro is far cheaper at $0.0408/run (roughly a sixth of Mini's price), and it's also 720p with native audio. What you pay the Mini premium for is the newer Seedance 2.0 architecture: steadier motion, better frame-to-frame stability, and two extra seconds (14s vs 12s). If budget is the deciding factor, 1.5 Pro is the honest pick; choose 2.0 Mini only when the motion quality of the newer generation is worth ~6x the per-run cost.
Seedance 2.0 Mini vs Veo 3
Both generate native audio. Veo 3 caps at 8 seconds per generation; Seedance 2.0 Mini runs to 14s in a single clip at roughly half the price ($0.2365 vs $0.48/run). Veo 3's edge is resolution — up to 1080p versus Mini's 720p. For longer single takes at lower cost, Mini wins; for sharper, audio-driven short shots, Veo 3 holds up.
Switching to Seedance 2.0 Mini from Dreamina / Volcano Engine
Output from the Seedance 2.0 Mini API on GPTProto is watermark-free and cleared for commercial use — the ads, product video, and social content you generate from your own inputs are yours to ship, with no overlay to clean up in post.
The model still applies its own safety and content filters at generation time, so some prompts — real public figures, copyrighted characters, and similar — may be refused. That's enforced inside the model: there's no way to route around a model's safety system, and we don't offer one. What you get is honest access: a developer API key, transparent per-run pricing, and clean, usable output you can take straight to a client.








