Text to Video
Write a description, get a finished clip. Official Seedance 2.5 and 2.0, up to 30 seconds at 1080P.
Pay as you go, with the exact cost shown before you generate
Describe a scene in plain language and the model returns a finished clip with camera movement and lighting. Seedance 2.5 runs to 30 seconds and can generate from audio alone; Seedance 2.0 tops out at 15 seconds but reaches 4K. Both generate sound with the video, quote the exact credit cost before the job is submitted, and refund it automatically if the generation fails.
- 01
Describe the scene, the camera movement and the mood
- 02
Pick a model, aspect ratio and duration, then check the credit cost
- 03
Submit and track progress under My Creations
Real generated samples
Every clip below was generated on this platform. The prompt is reproduced as written.
Majestic snow-capped peaks above a flowing sea of clouds at sunrise, cinematic aerial shot.
Retro gaming products on a dark desk, premium commercial motion and controlled studio light.
Aerial view of a modern city at golden hour, cinematic movement and photorealistic detail.
Cost per second
1 credit = $0.01. The table shows 480P. Higher resolutions and longer clips cost more, and the exact figure is shown before you generate.
| Model | Per second | Max length | Resolutions |
|---|---|---|---|
| Seedance 2.5 | $0.165 | 30s | 480P / 720P / 1080P |
| Seedance 2.0 | $0.113 | 15s | 480P / 720P / 1080P / 4K |
| Seedance 2.0 Fast | $0.075 | 15s | 480P / 720P |
| Seedance 2.0 Mini | $0.058 | 15s | 480P / 720P |
In depth
Vertical 9:16 story shots for TikTok, Reels and Shorts
Short-form story videos are built shot by shot, not in one prompt. Open on the conflict, because the first second decides whether the clip gets watched, and give each clip one beat with one camera move: a push-in on a reaction, a whip-pan to the reveal. A beat typically fits in five to fifteen seconds; a whole storyline packed into a single prompt tends to come back as a rushed montage.
Seedance generates sound together with the picture, speech included. When a character talks, name the speaker before the line and put the line in quotes; it helps the voice land on the right face. If the same character has to appear in every episode, move the series to reference mode and attach the character, rather than re-describing them in each prompt.
What a prompt needs to contain
Text-to-video has no reference image to anchor it, so everything the model needs has to come from the prompt. Prompts that read like a shot list outperform prompts that read like a wish. Four things carry most of the weight:
- Subject and action
- Who or what, doing what. "A red fox" is a subject; "a red fox stepping through wet grass" is a subject with motion the model can render over time.
- Camera
- Static, slow push-in, orbit, handheld. Without a camera instruction the model picks one, and it may change mid-clip.
- Light and time
- Overcast noon, low golden sun, single practical lamp. Lighting is what makes an eight-second clip look deliberate rather than generic.
- Rendering style
- Photographic, anime cel, claymation, 16mm film. Naming a medium constrains the whole frame more effectively than adjectives like "beautiful".
Choosing duration and frame
Seedance 2.5 generates 4 to 30 seconds in one pass. Longer is not automatically better: a single prompt describing one continuous action holds together across the full length, while a prompt containing three separate beats tends to produce three half-finished ones. For a multi-shot sequence, generate the shots separately and cut them together.
Aspect ratios available: 16:9, 9:16, 1:1, 4:3, 3:4, 21:9. There is also an adaptive setting that lets the model choose. Pick the frame before writing the prompt — a composition described for 16:9 rarely survives being rendered at 9:16.
When a prompt is refused
Prompt-level refusals name the prompt rather than any input file, and they are the only failure class in text-to-video that you can fix by rewriting. Rewording the specific phrase usually clears it. A prompt refused on Seedance 2.5 sometimes passes on Seedance 2.0 — the two models do not apply identical thresholds. Credits frozen for a refused job are released automatically.
Questions
- How long can a text-to-video clip be?
- Seedance 2.5 generates 4 to 30 seconds in a single pass. A prompt describing one continuous action holds together across the full length; for a multi-shot sequence, generate the shots separately and cut them together.
- Why was my prompt refused?
- Prompt-level refusals point at the wording, not at any input file, and rewriting the specific phrase usually clears it. A prompt refused on Seedance 2.5 sometimes passes on Seedance 2.0. Credits frozen for a refused job are released automatically.
- Can a text-to-video clip include dialogue?
- Yes. Seedance generates audio together with the picture, including speech. Name the speaker before each line and put the line in quotes. For a character who has to look the same in every episode, use reference mode instead.
- How are credits consumed?
- All video generations consume credits based on the selected model, duration, and quality. You can see the estimated cost before generating and top up credits with Stripe when needed.
- Are credits refunded when generation fails?
- Yes. If a generation fails because of a system, provider, or task error, frozen credits are released automatically. Review the task in My Creations and contact support if anything looks incorrect.
- Can I use the videos commercially?
- Yes, you own the commercial rights to the videos generated via AiuniVid. You can use them for social media, ads, YouTube, etc. Please refer to our Terms of Service for details.
- How do the Seedance versions differ?
- Seedance 2.5 supports omni references, audio, and clips up to 30 seconds. Seedance 2.0 is best for high-quality visuals and higher resolutions. Fast is quicker, and Mini is lower-cost for drafts and batch tests.
- What if generation fails?
- If generation fails due to a system error, credits are refunded automatically. For help, message us on Telegram @Jarvis11x or email [email protected].
