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

# Get Request Usage

> Look up a VoidAI request by completion ID or request UUID and inspect recorded tokens, credits, status, and latency.

`GET https://api.voidai.app/v1/usage/{id}`

Requires [authentication](/authentication). Supply a completion/response ID or the internal request UUID. The authenticated account must own the request; an ID is not an access credential.

```bash theme={null}
curl "https://api.voidai.app/v1/usage/$VOIDAI_RESPONSE_ID" \
  -H "Authorization: Bearer $VOIDAI_API_KEY"
```

| Field | Meaning |
| - | - |
| `id` | Completion ID, or request ID when no completion ID is recorded. |
| `status`, `model` | Recorded request status and model. |
| `usage.total_tokens`, `usage.input_tokens`, `usage.output_tokens` | Recorded token counts. |
| `credits_used` | Recorded charge in credits. |
| `latency_ms` | Recorded latency in milliseconds. |
| `created_at`, `completed_at` | Unix timestamps in milliseconds; completion can be null. |

Additional routing metadata may be returned; integrations should not depend on it. An unfinished request may not yet have final usage. See [billing](/guides/credits) for how cache and model rates affect charges, and [credits](/api-reference/credits/get) for the account balance.