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

# Models

> Discover the video models available through the Protoface API.

The model catalog tells you which video models are available and what each one
accepts. Read it at runtime instead of keeping a fixed list in your application.

## List models

```sh theme={null}
curl https://api.protoface.com/v1/models \
  -H "Authorization: Bearer $PROTOFACE_API_KEY"
```

Each model includes:

* `id`: the value used in `POST /v1/run/{model_id}`.
* `operations`: supported operations and their input fields.
* `input_modalities` and `output_modalities`: accepted and returned media.
* `pricing`: current credit pricing.
* `request_schemas`: JSON Schema for the request, when available.
* `revision` and `contract_revision`: versions of the model and its API contract.

## Retrieve one model

```sh theme={null}
curl https://api.protoface.com/v1/models/minimax-h3 \
  -H "Authorization: Bearer $PROTOFACE_API_KEY"
```

Use `operations` and `request_schemas` to build the request. Models can differ
in duration, quality, aspect ratio, and supported reference media.

## Submit to a model

Put the model ID in the URL:

```sh theme={null}
curl -X POST https://api.protoface.com/v1/run/minimax-h3 \
  -H "Authorization: Bearer $PROTOFACE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "prompt": "A close shot of rain moving across a window at night.",
    "duration_seconds": 8,
    "quality": "768p",
    "aspect_ratio": "16:9"
  }'
```

If a model is unavailable or the request is invalid for that model, the API
rejects the run before reserving credits.

## Next

<Columns cols={2}>
  <Card title="Model inference" href="/guides/model-inference" icon="boxes">
    Run your first model.
  </Card>

  <Card title="Runs" href="/guides/runs" icon="list-checks">
    Monitor and retrieve inference results.
  </Card>
</Columns>