> ## Documentation Index
> Fetch the complete documentation index at: https://docs.laozhang.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Muse Video API: 60-Second Clips and Pricing

> Generate one continuous 40-, 50-, or 60-second AI video with Muse on LaoZhang API at a flat price per clip, and see how continuation now counts the source.

Muse on LaoZhang API now makes a single continuous clip of up to 60 seconds, priced per video from \$1.05 to \$1.55. Continuation changed on the same day: the model name now sets the length of the finished video, source clip included.

* **Published**: November 2, 2026
* **Last checked**: November 2, 2026
* **Status**: Active

## Key facts

| Item | Details |
| - | - |
| New model IDs | `muse-video-40s`, `muse-video-50s`, `muse-video-60s` |
| Create a task | `POST /v1/videos` |
| Get a task | `GET /v1/videos/{id}` |
| Download | `video_url` in the finished task, or `GET /v1/videos/{id}/content` |
| Token | `default`, `vip`, or `svip` group; Usage first (recommended) or Per-call billing mode |
| Price | Per video; see [Pricing](#pricing) |
| Output | 720p-class MP4 (1280×720 at 16:9), 24 fps, with music and sound effects; links kept 7 days |
| Continuation | Send `video_url`; the finished video is as long as the model name says |
| Generation time | About 5 min for 40 s, about 7.5 min for 50 s, under 10 min for 60 s |
| Route | Meta's official API has no video generation model yet; this route is integrated independently and priced on its own |

## Pricing

| Model | Clip length | Per video |
| - | - | -: |
| `muse-video-40s` | 40 s | \$1.05 |
| `muse-video-50s` | 50 s | \$1.30 |
| `muse-video-60s` | 60 s | \$1.55 |

A 10-second clip costs \$0.30, and every extra 10 seconds adds \$0.25, so a 60-second clip works out to about \$0.026 per second. The existing models keep their prices: `muse-video` at \$0.30, `muse-video-20s` at \$0.55, and `muse-video-30s` at \$0.80.

* Failed tasks are refunded in full, and requests rejected at submission are not charged.
* Submitting the same request again is a new task and is charged again.
* See [Pricing in the Muse video guide](https://docs.laozhang.ai/en/api-capabilities/muse-video-generation.md#pricing) for every model and rule.

## Who is affected

**Affected**

* **Anyone who needs a clip longer than 30 seconds**: the three new models take the same request format as `muse-video-20s`, and existing tokens in the standard groups work as they are.
* **Continuation users on `muse-video-20s` and `muse-video-30s`**: since November 2, 2026, the model name sets the total length, source included. A 5-second source continued with `muse-video-20s` used to give a 15-second video and now gives a 20-second one. Pick the model by the total length you want.

**Not affected**

* **`muse-video`**: 5-, 8-, and 10-second clips keep the same price and behavior. It does not accept `video_url`.
* **Text- and image-to-video on `muse-video-20s` and `muse-video-30s`**: clip length and price are unchanged.
* **Other video models**: MiniMax H3, Seedance, and Wan keep their prices and request formats.
* **Tokens set to Usage-based only**: still cannot call Muse. Switch the billing mode to Usage first or Per-call.

## Generate a 60-second video

**1. Prepare a token**

In [Token management](https://api.laozhang.ai/token), use or create a token in the `default` group with the Usage first billing mode, then export it as `LAOZHANG_API_KEY`.

**2. Create a 60-second task**

```bash theme={null}
curl -s -X POST "https://api.laozhang.ai/v1/videos" \
  -H "Authorization: Bearer $LAOZHANG_API_KEY" \
  -H "Content-Type: application/json" \
  --data-raw '{
    "model": "muse-video-60s",
    "prompt": "A sailboat leaves a quiet harbor at sunrise, crosses open water as the light turns golden, and reaches a rocky island by midday, smooth tracking shot, cinematic lighting",
    "aspect_ratio": "16:9"
  }'
```

The response contains the task `id`; save it with `export VIDEO_ID=<task id>`. You can leave out `seconds`; if you send it, it must be `60` for this model.

**3. Poll and download**

```bash theme={null}
curl -s "https://api.laozhang.ai/v1/videos/$VIDEO_ID" \
  -H "Authorization: Bearer $LAOZHANG_API_KEY"
```

Poll every 15 seconds. When `status` is `completed`, download the MP4 from `video_url`, or call `GET /v1/videos/{id}/content`. Video links stay available for 7 days.

To extend an existing clip to 40 seconds in total, send its link as `video_url`:

```bash theme={null}
curl -s -X POST "https://api.laozhang.ai/v1/videos" \
  -H "Authorization: Bearer $LAOZHANG_API_KEY" \
  -H "Content-Type: application/json" \
  --data-raw '{
    "model": "muse-video-40s",
    "prompt": "The boat drifts into the morning mist as the sun rises over the sea",
    "video_url": "https://example.com/boat-10s.mp4"
  }'
```

This turns a 10-second source into a 40-second video. For the full continuation rules, see [continuation in the Muse video guide](https://docs.laozhang.ai/en/api-capabilities/muse-video-generation.md#make-20s-30s-and-up-to-60s-clips-or-continue-a-video).

## FAQ

### How long does a 60-second clip take to generate?

Usually under 10 minutes. A 40-second clip takes about 5 minutes, a 50-second clip about 7.5 minutes, and clips up to 10 seconds take 2–3 minutes.

The API is asynchronous, so keep polling for up to 30 minutes and do not resubmit while the task runs, or you will be charged again. A task still unfinished after 40 minutes is marked failed and refunded.

### What source video can I continue?

**The source must be at least 2 seconds shorter than the finished video.**

* `muse-video-40s`: up to 38 seconds
* `muse-video-50s`: up to 48 seconds
* `muse-video-60s`: up to 58 seconds
* `muse-video-20s` and `muse-video-30s`: up to 18 and 28 seconds

A source file larger than about 19 MB is first compressed to 720p, so that part of the result looks slightly softer. A `video_url` returned by LaoZhang API works as a source.

### How long can other LaoZhang API video models go?

Muse is the only video model on LaoZhang API that generates 60 seconds in one task. MiniMax H3 goes up to 15 seconds, Seedance 2.5 and Wan 3.0 up to 30 seconds, and the Seedance 2.0 series up to 15 seconds. Compare cost and features in the [video generation API comparison](https://docs.laozhang.ai/en/api-capabilities/video-generation-guide.md).

### Can I get 1080p from the 60-second model?

No. All Muse models output 1280×720 in landscape and 720×1280 in portrait, at 24 fps.

## Sources and related links

* [Meta Muse Video API guide](https://docs.laozhang.ai/en/api-capabilities/muse-video-generation.md): request examples, fields, task status, and common errors.
* [Video generation API comparison](https://docs.laozhang.ai/en/api-capabilities/video-generation-guide.md): price, length, and features across video models.
* [Muse video launch notice](https://docs.laozhang.ai/en/announcements/muse-video-2026-10.md): the October 30, 2026 launch of `muse-video`, `muse-video-20s`, and `muse-video-30s`.
* [Meta Model API models](https://dev.meta.ai/resources/blog/build-with-muse-spark): the models on Meta's official API, with no video generation model (checked October 30, 2026).
* [Models and pricing](https://docs.laozhang.ai/en/models.md)
* [LaoZhang API announcements](https://docs.laozhang.ai/en/changelog.md)


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.