Feat: Let providers configure a fast model for summarization (#4228)

crates/goose/src/providers/base.rs

@@ -317,25 +317,40 @@ pub trait Provider: Send + Sync {
     where
         Self: Sized;
 
-    /// Generate the next message using the configured model and other parameters
-    ///
-    /// # Arguments
-    /// * `system` - The system prompt that guides the model's behavior
-    /// * `messages` - The conversation history as a sequence of messages
-    /// * `tools` - Optional list of tools the model can use
-    ///
-    /// # Returns
-    /// A tuple containing the model's response message and provider usage statistics
-    ///
-    /// # Errors
-    /// ProviderError
-    ///   - It's important to raise ContextLengthExceeded correctly since agent handles it
+    // Internal implementation of complete, used by complete_fast and complete
+    // Providers should override this to implement their actual completion logic
+    async fn complete_with_model(
+        &self,
+        model_config: &ModelConfig,
+        system: &str,
+        messages: &[Message],
+        tools: &[Tool],
+    ) -> Result<(Message, ProviderUsage), ProviderError>;
+
+    // Default implementation: use the provider's configured model
     async fn complete(
         &self,
         system: &str,
         messages: &[Message],
         tools: &[Tool],
-    ) -> Result<(Message, ProviderUsage), ProviderError>;
+    ) -> Result<(Message, ProviderUsage), ProviderError> {
+        let model_config = self.get_model_config();
+        self.complete_with_model(&model_config, system, messages, tools)
+            .await
+    }
+
+    // Check if a fast model is configured, otherwise fall back to regular model
+    async fn complete_fast(
+        &self,
+        system: &str,
+        messages: &[Message],
+        tools: &[Tool],
+    ) -> Result<(Message, ProviderUsage), ProviderError> {
+        let model_config = self.get_model_config();
+        let fast_config = model_config.use_fast_model();
+        self.complete_with_model(&fast_config, system, messages, tools)
+            .await
+    }
 
     /// Get the model config from the provider
     fn get_model_config(&self) -> ModelConfig;
@@ -418,7 +433,7 @@ pub trait Provider: Send + Sync {
         let prompt = self.create_session_name_prompt(&context);
         let message = Message::user().with_text(&prompt);
         let result = self
-            .complete(
+            .complete_fast(
                 "Reply with only a description in four words or less",
                 &[message],
                 &[],