> ## 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.

# PrivateEndpointPromptCaching

Prompt-caching controls for an OpenAI-compatible private endpoint. Accepted fields default on after validation; explicit owner settings override the defaults.

## Example Usage

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

let value: PrivateEndpointPromptCaching = {
  cacheControl: false,
  promptCacheKey: false,
};
```

## Fields

| Field | Type | Required | Description | Example |
| - | - | - | - | - |
| `cacheControl` | *boolean* | ✔️ | Forward top-level, message, and tool cache\_control markers. Defaults on after endpoint validation accepts these fields; otherwise off. Set false to opt out, or true to declare support explicitly. | false |
| `promptCacheKey` | *boolean* | ✔️ | Send an OpenRouter-derived, customer-isolated prompt\_cache\_key, not the caller-supplied key. Defaults on after endpoint validation accepts the field; otherwise off. Set false to opt out, or true to declare support explicitly. HIPAA and unknown workspace posture withhold OpenRouter caching directives and the derived key; this does not control upstream retention. | false |
