> For clean Markdown of any page, append .md to the page URL.
> For a complete documentation index, see https://docs.reka.ai/vision/errors/llms.txt.
> For AI client integration (Claude Code, Cursor, etc.), connect to the MCP server at https://docs.reka.ai/_mcp/server.

# Errors

> **Info**
>
> For chat over text, images and short videos, use the OpenAI-compatible [Chat Completions API](/chat/overview).

# Vision API errors

Errors come back as JSON in this shape:

```json
{
  "error": {
    "type": "not_found_error",
    "message": "Video not found",
    "param": "video_id",
    "details": [...]
  }
}
```

| Field                | Type   | Description                                               |
| -------------------- | ------ | --------------------------------------------------------- |
| `type`               | string | Error category for client-side handling (see table below) |
| `message`            | string | Human-readable explanation                                |
| `param` (optional)   | string | When present, the request parameter that caused the error |
| `details` (optional) | array  | When present, more detail on the request error            |

## Error types and HTTP status codes

| Status | Error Type             | Client Action                                                                                   |
| ------ | ---------------------- | ----------------------------------------------------------------------------------------------- |
| 400    | `validation_error`     | Fix your request and retry                                                                      |
| 401    | `authentication_error` | Check that you're using a valid Reka API key                                                    |
| 404    | `not_found_error`      | Resource not found                                                                              |
| 429    | `rate_limit_error`     | Wait until your rate limit is restored and retry, or upgrade your API tier for higher limits    |
| 500    | `server_error`         | Retry with backoff. If this persists, contact Reka at [contact@reka.ai](mailto:contact@reka.ai) |

## Request ID

Every response includes an `X-Request-ID` header for log correlation:

```
X-Request-ID: a1b2c3d4e5f6...
```

The response body does not repeat it. Include this ID when you report an issue.

> One OpenAI-compatible API for Reka models and a curated selection of open models.