How to write prompts for Seedance 2.0 Mini
Seedance 2.0 Mini turns words or a start picture into a clip, with sound. The first two sections put the provider’s own guide in our words, and we did not test them. The last rests on six sample clips made with this model that we watched ourselves, a small sample.
Written from the provider’s documentation and from results we checked ourselves. Last checked November 8, 2026.
Make something with Seedance 2.0 MiniThe Seedance 2.0 Mini model page
What the guide says about the 2.0 series
From the provider’s documentation
- BytePlus names three models in the Seedance 2.0 series: Seedance 2.0, Seedance 2.0 Fast and Seedance 2.0 Mini. It says they support largely the same features and differ mainly in generation quality and cost, and that only Seedance 2.0 outputs 4K.
- The guide is written for the whole series, and much of it covers reference pictures, video and audio clips, and editing or extending a video. On Pickshot this model takes words or a start picture, so those parts do not apply here.
- Its basic idea is that a good prompt is an instruction to a director, not copywriting. Say who does what, then where and in what mood, then how to shoot it, and tighten the result with style, picture quality and constraint words.
- Keep the description short and without contradictions, and use the same word for the same subject every time you mention it.
Shots, actions and constraint words
From the provider’s documentation
- For anything complex, write a storyboard: Shot 1, Shot 2, Shot 3, each with the camera move or the way the shot changes, the subject’s action and expression, where it happens and the sound. The guide warns that exact timing such as 0 to 3 seconds is unreliable and that forcing a length can give odd results.
- Describe actions by body part and degree, such as a hand raised slowly or a head turned quickly. The guide prefers slow, gentle, continuous movements to sprinting, big jumps or rolls, and wants one motion to follow the previous one naturally. Show feeling through physical detail rather than words such as very sad.
- Use standard camera terms, and ask for one kind of camera move per shot. Asking for a push, a pull and a pan at once makes the picture unstable.
- Add constraint words at the end, such as keep the clip without subtitles, do not generate a logo, do not generate a watermark. The guide says subtitles cannot be ruled out completely and that landscape clips show them less often than portrait ones. If you add speech, write it in one language.
- To keep a 2D or 3D animation look, name the style in the prompt. A realistic start picture can otherwise drift toward live action.
What our sample clips showed
From results we checked ourselves
- The puppy clip shows a shiba puppy bounding through fresh snow toward the camera, with snow puffing up in a bright, backlit morning. The prompt said ears flopping, and the ears stay upright.
- The Seongsu clip starts as a close selfie, then she steps back, walks toward the camera, smiles and waves, in golden light with a cafe terrace behind her. The cafe sign shows partly visible English letters, text the prompt did not ask for.
- The courtyard clip shows a woman twirling in a sheer white hanbok-style dress with petals falling in front of palace roofs, as described.
- The maple clip shows leaves falling past a temple gate while the camera moves down from the branches to a moss-covered stone lantern. The prompt named no camera move, so that descent was the model’s choice.
- The latte clip opens on a medium shot of the barista pouring, then cuts to a close-up of the cup as the heart forms. The prompt asked for the close-up, and the clip reached it with a cut part-way through.
- The whale clip shows a pale whale with glowing patterns on its skin swimming across a night sky above stilt houses on still water. We did not judge the sound of any clip, and the prompts were single sentences, so these clips say nothing about reference pictures.
Real results and the prompts behind them
Results made with this model. The prompts are shown exactly as written, in English.
Settings of this model
- Van
- ByteDance
- Typ
- AI video model
- Manier van maken
- Tekst naar video · Afbeelding naar video
- Lengte
- 4–15s
- Resolutie
- 480p / 720p
- Aspect ratios
- 16:9 · 9:16 · 1:1 · 4:3 · 3:4
- Geluid
- Met geluid
- Prijs
- from 61 / sec
Vragen
Can I use reference pictures, clips or audio with Seedance 2.0 Mini?
The provider’s guide covers them for the 2.0 series, with tags such as Image 1 in the prompt. On Pickshot this model takes words or a start picture, so we have not tried those parts.
How do I stop subtitles appearing?
The guide suggests writing that the clip must have no subtitles or text. It says this lowers the chance but cannot remove it, and that landscape clips show subtitles less often than portrait ones. We did not test it.
Should I give exact seconds for each shot?
The guide says exact timing such as 0 to 3 seconds is unreliable. Write Shot 1, Shot 2 and Shot 3 in order and let the model set the pace.
Provider documentation we read
We read these pages and wrote the points that matter in our own words. The provider may change them.
- Dreamina Seedance 2.0 series prompt guide (BytePlus ModelArk) (docs.byteplus.com)
- Dreamina Seedance 2.0 series tutorial (BytePlus ModelArk) (docs.byteplus.com)
More prompting guides
- How to write prompts for GPT Image 2.6 FlareAI image model
- How to write prompts for GPT Image 2.6 SunburstAI image model
- How to write prompts for Nano Banana 2.1AI image model
- How to write prompts for MOSS ImageAI image model
- How to write prompts for Grok Imagine ProAI image model
- How to write prompts for Nano Banana ProAI image model
- How to write prompts for Nano Banana 2AI image model
- How to write prompts for GPT Image 2.5 FlareAI image model
- How to write prompts for IndigoAI image model
- How to write prompts for SaffronAI image model
- How to write prompts for CeladonAI image model
- How to write prompts for LapisAI image model
- How to write edit requests for UmberAI-afbeeldingseditor
- How to write prompts for Kling Video 4.0AI video model
- How to write prompts for Seedance 2.5AI video model
- How to write prompts for Kling V3AI video model
- How to write prompts for Kling O3 OmniAI video model
- How to write prompts for Veo 3.1AI video model
- How to write prompts for Veo 3.1 FastAI video model
- How to write prompts for SilkAI video model
- How to write prompts for LacquerAI video model
- How to write prompts for SatinAI video model
De namen van modellen, bedrijven en producten op deze site zijn handelsmerken van hun eigenaren; Pickshot is niet verbonden aan, gesponsord door of goedgekeurd door hen.