Seedance 2 Fast is the production-oriented option for teams that want quicker and lower-cost iteration across text-to-video and image-to-video. The MakeInfluencer contract supports 4–15 second clips, 480p through 1080p, generated audio, an optional end frame, and reference images.

## Model Contract

| Field              | Behavior                             |
| ------------------ | ------------------------------------ |
| `prompt`           | Required scene and motion direction  |
| `startFrameUrl`    | Optional; switches to image-to-video |
| `endFrameUrl`      | Optional last-frame guidance         |
| `aspectRatio`      | Defaults to 16:9 for text generation |
| `duration`         | 4–15 seconds, default 5              |
| `resolution`       | `480p`, `720p`, or `1080p`           |
| `generateAudio`    | Defaults to true                     |
| `reference_images` | Optional reference image array       |

The [Seedance 2 Fast API reference](/developers/models/seedance-2-fast-api) reflects the live registry and exact credit quote.

## Example Request

```bash
curl --request POST \
  https://www.makeinfluencer.ai/api/v1/models/seedance2fast/preview-credits \
  --header "Authorization: Bearer $MAKEINFLUENCER_API_KEY" \
  --header "Content-Type: application/json" \
  --data '{
    "input": {
      "prompt": "A creator turns toward camera on a busy night street, slow push-in, natural traffic ambience",
      "startFrameUrl": "https://example.com/approved-creator.jpg",
      "duration": 5,
      "resolution": "720p",
      "generateAudio": true
    }
  }'
```

The preview accounts for duration, resolution, and whether references are present. That is important because reference-assisted generation can use a different provider rate.

## When to Use 480p, 720p, or 1080p

### 480p for Iteration

Use 480p to test motion prompts, camera behavior, and whether the source image animates cleanly. A draft that fails creatively does not become better because it was rendered at 1080p.

### 720p for Working Content

Use 720p for internal review, fast social experiments, and workflows where output speed and cost matter.

### 1080p for Approved Directions

Use 1080p after the prompt, source image, and motion have already succeeded. Preview the exact credit difference before promoting a batch.

<Callout type="tip" title="Escalate Resolution, Not Uncertainty">
  Find the winning motion at the lowest acceptable resolution. Re-render only
  approved directions at the delivery quality your customer needs.
</Callout>

## Start and End Frame Workflow

An end frame can guide where a shot should arrive, but it also creates a harder constraint. Use it when the final composition matters, such as:

- A product reveal that ends on a pack shot
- A character moving from a wide frame to a close-up
- A transition between two approved campaign poses
- A loop designed to return toward a known visual state

Keep the start and end images compatible in identity, clothing, lighting, and scene geometry. Large contradictions force the model to invent an implausible transition.

## Generated Audio vs Controlled Speech

Set `generateAudio` when ambient sound or model-generated audio benefits the clip. For exact spoken content, use a separate audio-driven workflow such as the [InfiniteTalk API](/developers/models/infinitetalk-api), where the uploaded audio controls timing and speech.

## Seedance 2 Fast vs Seedance 2.5

Seedance 2 Fast is live and available through the current registry. Seedance 2.5 has been unveiled, but production API access is still being verified. Track the [Seedance 2.5 API availability page](/guides/seedance-2-5-api-availability) for the integration status.

Do not block a product launch on an unverified future endpoint. Keep the model ID configurable, ship on a production model, and compare 2.5 when its pricing and reliability are real.

## API FAQ

### Does Seedance 2 Fast support text-to-video?

Yes. Omit `startFrameUrl` and send a prompt plus the desired aspect ratio, duration, and resolution.

### Does it support image-to-video?

Yes. Add `startFrameUrl`; you can also provide an end frame where supported by the model contract.

### Does reference usage affect price?

It can. The registry has separate per-second pricing tables with and without references. Always call `preview-credits` with the exact planned input.

### Is Seedance 2.5 already available?

Not through MakeInfluencer as of August 5, 2026. The planned integration will happen after production provider access and verification.


Use the Seedance 2 Fast API for text-to-video or image-to-video clips from 4–15 seconds, 480p–1080p output, optional references, end frames, and audio.

Seedance 2 Fast API Guide: Faster Video with Audio

Does Seedance 2 Fast support text-to-video?

Seedance 2 Fast API Guide: Faster Video with Audio

Create Your AI Influencer

Continue Reading

Best Civitai Alternative for Hosted AI Creation

Faceless Video API vs AI Model API: Which Workflow Fits?

Getting Started with MakeInfluencer.ai

Ready to try it yourself?