Response Headers — Markdown source

> **AI Agents**: Index `/api/llms.txt` | Full `/api/llms-full-en.txt` | OpenAPI `/api/openapi.yaml`
> Base URL: `<BASE_URL>/v1`

# Response Headers

X6.AI adds diagnostic headers on API responses.

## Global headers

| Header | Present on | Description |
|--------|------------|-------------|
| `X-SNX-Trace-ID` | All responses | Unique request id; include in support tickets |
| `X-Request-Id` | All responses | Same value as `X-SNX-Trace-ID`; use with [request reconciliation](./billing-requests.md) |

## Conversation headers

| Header | Present on | Description |
|--------|------------|-------------|
| `X-SNX-Model` | `POST /v1/chat/completions`, `/v1/messages`, `/v1/responses` | Logical model id requested |
| `X-SNX-Provider` | Same | Vendor identifier for the inference provider |

Example:

```http
HTTP/1.1 200 OK
Content-Type: application/json
X-SNX-Trace-ID: 7c9e6679-7425-40de-944b-e07fc1f90ae7
X-Request-Id: 7c9e6679-7425-40de-944b-e07fc1f90ae7
X-SNX-Model: deepseek/deepseek-chat
X-SNX-Provider: deepseek
```

## Quota responses

When a usage quota limit applies, X6.AI may include:

```http
Retry-After: 3600
```

Value is seconds until retry is suggested.

## CORS

X6.AI allows these request headers from browsers:

- `Authorization`
- `Content-Type`
- `Accept-Language`
- `X-Locale`

## Idempotency

`Idempotency-Key` is a **request** header (not a response header). See [Chat Completions](./chat-completions.md).

## Related

- [Chat Completions](./chat-completions.md)
- [Errors](./errors.md)