🪙 fix: Use Fallback Token Transaction if No Collected Usage (#8503)

api/app/clients/BaseClient.js

@@ -108,12 +108,15 @@ class BaseClient {
   /**
    * Abstract method to record token usage. Subclasses must implement this method.
    * If a correction to the token usage is needed, the method should return an object with the corrected token counts.
+   * Should only be used if `recordCollectedUsage` was not used instead.
+   * @param {string} [model]
    * @param {number} promptTokens
    * @param {number} completionTokens
    * @returns {Promise<void>}
    */
-  async recordTokenUsage({ promptTokens, completionTokens }) {
+  async recordTokenUsage({ model, promptTokens, completionTokens }) {
     logger.debug('[BaseClient] `recordTokenUsage` not implemented.', {
+      model,
       promptTokens,
       completionTokens,
     });
@@ -741,9 +744,13 @@ class BaseClient {
       } else {
         responseMessage.tokenCount = this.getTokenCountForResponse(responseMessage);
         completionTokens = responseMessage.tokenCount;
+        await this.recordTokenUsage({
+          usage,
+          promptTokens,
+          completionTokens,
+          model: responseMessage.model,
+        });
       }
-
-      await this.recordTokenUsage({ promptTokens, completionTokens, usage });
     }
 
     if (userMessagePromise) {

api/server/controllers/agents/client.js

@@ -1135,8 +1135,52 @@ class AgentClient extends BaseClient {
     }
   }
 
-  /** Silent method, as `recordCollectedUsage` is used instead */
-  async recordTokenUsage() {}
+  /**
+   * @param {object} params
+   * @param {number} params.promptTokens
+   * @param {number} params.completionTokens
+   * @param {OpenAIUsageMetadata} [params.usage]
+   * @param {string} [params.model]
+   * @param {string} [params.context='message']
+   * @returns {Promise<void>}
+   */
+  async recordTokenUsage({ model, promptTokens, completionTokens, usage, context = 'message' }) {
+    try {
+      await spendTokens(
+        {
+          model,
+          context,
+          conversationId: this.conversationId,
+          user: this.user ?? this.options.req.user?.id,
+          endpointTokenConfig: this.options.endpointTokenConfig,
+        },
+        { promptTokens, completionTokens },
+      );
+
+      if (
+        usage &&
+        typeof usage === 'object' &&
+        'reasoning_tokens' in usage &&
+        typeof usage.reasoning_tokens === 'number'
+      ) {
+        await spendTokens(
+          {
+            model,
+            context: 'reasoning',
+            conversationId: this.conversationId,
+            user: this.user ?? this.options.req.user?.id,
+            endpointTokenConfig: this.options.endpointTokenConfig,
+          },
+          { completionTokens: usage.reasoning_tokens },
+        );
+      }
+    } catch (error) {
+      logger.error(
+        '[api/server/controllers/agents/client.js #recordTokenUsage] Error recording token usage',
+        error,
+      );
+    }
+  }
 
   getEncoding() {
     return 'o200k_base';