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

# ProviderResponse

Details of a provider response for a generation attempt

## Example Usage

```typescript theme={null}
import { ProviderResponse } from "@openrouter/sdk/models";

let value: ProviderResponse = {
  status: 200,
};
```

## Fields

| Field | Type | Required | Description | Example |
| - | - | - | - | - |
| `endpointId` | *string* | :heavy\_minus\_sign: | Internal endpoint identifier | ep\_abc123 |
| `id` | *string* | :heavy\_minus\_sign: | Upstream provider response identifier | chatcmpl-abc123 |
| `isByok` | *boolean* | :heavy\_minus\_sign: | Whether the request used a bring-your-own-key | false |
| `latency` | *number* | :heavy\_minus\_sign: | Response latency in milliseconds | 1200 |
| `modelPermaslug` | *string* | :heavy\_minus\_sign: | Canonical model slug | openai/gpt-4 |
| `providerName` | [models.ProviderResponseProviderName](../models/providerresponseprovidername.mdx) | :heavy\_minus\_sign: | Name of the provider | OpenAI |
| `routedServiceTier` | [models.RoutedServiceTier](../models/routedservicetier.mdx) | :heavy\_minus\_sign: | The service tier this request was routed to (e.g. flex, priority). The tier actually applied and billed is determined by the provider response and may differ. | priority |
| `status` | *number* | :heavy\_check\_mark: | HTTP status code from the provider | 200 |