Product videos
Describe the product, where it will be used and the pacing of each shot. Review the generated details for accuracy before publishing.
Back to models
Integration guide
Video
Seedance 2.0 Fast
dreamina-seedance-2-0-fast-260128The Fast variant of Seedance 2.0, with 4–15 second output in 480p or 720p.
- Modality
- Video
- List price
- $5.60 / million tokens≈ $0.605 per 5s at 720p
- Video duration
- 4–15 seconds
Model specifications and parameters
Fields are shown for this model's connected endpoint, not one generic form for every model.
| Parameter / field | Current range | Description |
|---|---|---|
| duration | 4–15 s / -1 | Omitted API duration defaults to 5 seconds. |
| resolution | 480p / 720p | Output resolution varies by model. |
| ratio | 21:9 / 16:9 / 4:3 / 1:1 / 3:4 / 9:16 / adaptive | A fixed ratio maps to valid dimensions; smart mode lets the model decide. |
| output_format | mp4 | Output video container format. |
| generate_audio | true / false | Available only when this model supports audio output. |
| watermark | true / false | Whether this model lets you control the output watermark. |
| content.role=first_frame | Optional | Optional reference image; first-frame rules vary by model. |
| content.role=last_frame | Optional; requires a first frame | First and last frames are separate image inputs. Matching ratios work best; the last frame follows the first frame's canvas. |
| mode | omni_reference / omni_edit / omni_extend | The web UI offers reference, edit and extend according to this model. LINK42 converts or removes its mode field before the provider call. |
| content.role=reference_image | 0–9 | Omni reference images and first/last frames are mutually exclusive. |
| content.role=reference_video | 0–3 | Reference videos may be up to 15 seconds. Video input selects its own price tier. |
| content.role=reference_audio | 0–3 | Reference audio may be up to 15 seconds and needs an image or video too. |
| Web batch count | 1–8 | The experience center submits up to eight separately billed tasks, not one upstream n parameter. |
Provider limits and LINK42 web-test limits may differ. LINK42's current quote and actual settlement govern price; external docs are provided to verify model capabilities.
Test a result before integrating
Adjust parameters, review the estimate and validate your idea with an actual output.
Test parameters
dreamina-seedance-2-0-fast-260128Text / frames
This model accepts up to 9 images, 3 videos and 3 audio clips. Video input uses its own price tier. Library clip lengths are checked again by the provider.
Fixed seconds
720p
16:9
Generate audio
AI watermark
First-frame image
Sign in to upload a file or pick one from your library. First-frame original: JPG, PNG, or WebP, up to 100 MB. Files over 10 MB are optimized before upload; changing the asset recalculates the estimate.
This model supports first and last frames. Add the first frame to reveal the last-frame control.
Checking your session…
Output
Your first result starts here
Enter a prompt and start a test. Generated text, images or video will appear here.
Test history
Private to your account, including recent results and actual charges.
Explore the model and adjust parameters first. Sign in to restore your draft and run a paid test.
Overview
Seedance 2.0 Fast
The Fast variant of Seedance 2.0, with 4–15 second output in 480p or 720p. Tasks are submitted through BytePlus ModelArk and results become available after asynchronous processing.
From prompt to finished video
Describe the subject, action, setting and camera movement, then choose the duration, aspect ratio and resolution. Track progress after submission and review the finished video alongside its actual usage charges.
Use cases
Storyboards and concept tests
Use short clips to explore camera movement, lighting and pacing, then refine your prompt based on the results.
Examples and ideas
Examples help you prepare inputs. Evaluate quality with your own actual runs.
Product showcase prompt
Illustrative asset, not a verified model outputAn unbranded white ceramic cup against a plain white background. Soft side lighting. The camera slowly moves closer while the cup remains still. A clean composition with no people, captions or watermarks.
Frequently asked questions
Are test generations billed?
Yes. Sign in to use your account balance. An estimated amount is reserved before submission; the final charge is based on actual usage, and any unused reservation is released.
What if video generation fails?
Check the task error and usage records first. Failed or cancelled tasks are settled or released through the ledger. Avoid repeatedly creating new tasks.
Does the playground call a real model?
Yes. Submitting a paid test sends a request to BytePlus. The examples shown here are prompts, not claimed model outputs.
Pricing
These are standard model prices. Signed-in estimates account for your billing identity and parameters; final charges use actual usage.
| Pricing tier | Output price | With video input |
|---|---|---|
| 480p | $5.6/ million tokens | $3.3/ million tokens |
| 720p | $5.6/ million tokens | $3.3/ million tokens |
Parameter-based pricing · See pricing details
API
After validating the result, use the same model identifier in your application. Keep API keys on your server.
cURL
curl https://www.link42.ai/v1/videos -H "Authorization: Bearer $LINK42_API_KEY" -H "Content-Type: application/json" -d '{"model": "dreamina-seedance-2-0-fast-260128", "content": [{"type": "text", "text": "…"}], "duration": 5, "resolution": "720p"}'