UsageMetadata - add cachedContentTokenCount field

.changeset/angry-hotels-learn.md

@@ -0,0 +1,5 @@
+---
+"@google/generative-ai": minor
+---
+
+Add a `cachedContentTokenCount` field to the `UsageMetadata` interface returned by `generateContent` responses.

common/api-review/generative-ai.api.md

@@ -638,6 +638,7 @@ export interface ToolConfig {
 
 // @public
 export interface UsageMetadata {
+    cachedContentTokenCount?: number;
     candidatesTokenCount: number;
     promptTokenCount: number;
     totalTokenCount: number;

docs/reference/main/generative-ai.usagemetadata.cachedcontenttokencount.md

@@ -0,0 +1,13 @@
+<!-- Do not edit this file. It is automatically generated by API Documenter. -->
+
+[Home](./index.md) &gt; [@google/generative-ai](./generative-ai.md) &gt; [UsageMetadata](./generative-ai.usagemetadata.md) &gt; [cachedContentTokenCount](./generative-ai.usagemetadata.cachedcontenttokencount.md)
+
+## UsageMetadata.cachedContentTokenCount property
+
+Total token count in the cached part of the prompt, i.e. in the cached content.
+
+**Signature:**
+
+```typescript
+cachedContentTokenCount?: number;
+```

docs/reference/main/generative-ai.usagemetadata.md

@@ -16,6 +16,7 @@ export interface UsageMetadata
 
 |  Property | Modifiers | Type | Description |
 |  --- | --- | --- | --- |
+|  [cachedContentTokenCount?](./generative-ai.usagemetadata.cachedcontenttokencount.md) |  | number | _(Optional)_ Total token count in the cached part of the prompt, i.e. in the cached content. |
 |  [candidatesTokenCount](./generative-ai.usagemetadata.candidatestokencount.md) |  | number | Total number of tokens across the generated candidates. |
 |  [promptTokenCount](./generative-ai.usagemetadata.prompttokencount.md) |  | number | Number of tokens in the prompt. |
 |  [totalTokenCount](./generative-ai.usagemetadata.totaltokencount.md) |  | number | Total token count for the generation request (prompt + candidates). |

packages/main/types/responses.ts

@@ -98,6 +98,8 @@ export interface UsageMetadata {
   candidatesTokenCount: number;
   /** Total token count for the generation request (prompt + candidates). */
   totalTokenCount: number;
+  /** Total token count in the cached part of the prompt, i.e. in the cached content. */
+  cachedContentTokenCount?: number;
 }
 
 /**