feat(plan): support automatic model switching for Plan Mode (google-gemini#20240)

Author: jerop

docs/cli/plan-mode.md

@@ -27,6 +27,7 @@ implementation. It allows you to:
     - [Example: Allow git commands in Plan Mode](#example-allow-git-commands-in-plan-mode)
     - [Example: Enable research subagents in Plan Mode](#example-enable-research-subagents-in-plan-mode)
   - [Custom Plan Directory and Policies](#custom-plan-directory-and-policies)
+- [Automatic Model Routing](#automatic-model-routing)
 
 ## Enabling Plan Mode
 
@@ -242,6 +243,32 @@ modes = ["plan"]
 argsPattern = "\"file_path\":\"[^\"]+[\\\\/]+\\.gemini[\\\\/]+plans[\\\\/]+[\\w-]+\\.md\""
 ```
 
+## Automatic Model Routing
+
+When using an [**auto model**], Gemini CLI automatically optimizes [**model
+routing**] based on the current phase of your task:
+
+1.  **Planning Phase:** While in Plan Mode, the CLI routes requests to a
+    high-reasoning **Pro** model to ensure robust architectural decisions and
+    high-quality plans.
+2.  **Implementation Phase:** Once a plan is approved and you exit Plan Mode,
+    the CLI detects the existence of the approved plan and automatically
+    switches to a high-speed **Flash** model. This provides a faster, more
+    responsive experience during the implementation of the plan.
+
+This behavior is enabled by default to provide the best balance of quality and
+performance. You can disable this automatic switching in your settings:
+
+```json
+{
+  "general": {
+    "plan": {
+      "modelRouting": false
+    }
+  }
+}
+```
+
 [`list_directory`]: /docs/tools/file-system.md#1-list_directory-readfolder
 [`read_file`]: /docs/tools/file-system.md#2-read_file-readfile
 [`grep_search`]: /docs/tools/file-system.md#5-grep_search-searchtext
@@ -259,3 +286,5 @@ argsPattern = "\"file_path\":\"[^\"]+[\\\\/]+\\.gemini[\\\\/]+plans[\\\\/]+[\\w-
 [YOLO mode]: /docs/reference/configuration.md#command-line-arguments
 [`plan.toml`]:
   https://github.com/google-gemini/gemini-cli/blob/main/packages/core/src/policy/policies/plan.toml
+[auto model]: /docs/reference/configuration.md#model-settings
+[model routing]: /docs/cli/telemetry.md#model-routing

docs/cli/settings.md

@@ -29,6 +29,7 @@ they appear in the UI.
 | Enable Auto Update      | `general.enableAutoUpdate`         | Enable automatic updates.                                                                                                                                                      | `true`      |
 | Enable Notifications    | `general.enableNotifications`      | Enable run-event notifications for action-required prompts and session completion. Currently macOS only.                                                                       | `false`     |
 | Plan Directory          | `general.plan.directory`           | The directory where planning artifacts are stored. If not specified, defaults to the system temporary directory.                                                               | `undefined` |
+| Plan Model Routing      | `general.plan.modelRouting`        | Automatically switch between Pro and Flash models based on Plan Mode status. Uses Pro for the planning phase and Flash for the implementation phase.                           | `true`      |
 | Max Chat Model Attempts | `general.maxAttempts`              | Maximum number of attempts for requests to the main chat model. Cannot exceed 10.                                                                                              | `10`        |
 | Debug Keystroke Logging | `general.debugKeystrokeLogging`    | Enable debug logging of keystrokes to the console.                                                                                                                             | `false`     |
 | Enable Session Cleanup  | `general.sessionRetention.enabled` | Enable automatic session cleanup                                                                                                                                               | `false`     |

docs/cli/telemetry.md

@@ -487,6 +487,7 @@ Captures Gemini API requests, responses, and errors.
     - `reasoning` (string, optional)
     - `failed` (boolean)
     - `error_message` (string, optional)
+    - `approval_mode` (string)
 
 #### Chat and streaming
 
@@ -711,12 +712,14 @@ Routing latency/failures and slash-command selections.
   - **Attributes**:
     - `routing.decision_model` (string)
     - `routing.decision_source` (string)
+    - `routing.approval_mode` (string)
 
 - `gemini_cli.model_routing.failure.count` (Counter, Int): Counts model routing
   failures.
   - **Attributes**:
     - `routing.decision_source` (string)
     - `routing.error_message` (string)
+    - `routing.approval_mode` (string)
 
 ##### Agent runs
 

docs/reference/configuration.md

@@ -137,6 +137,12 @@ their corresponding top-level category object in your `settings.json` file.
   - **Default:** `undefined`
   - **Requires restart:** Yes
 
+- **`general.plan.modelRouting`** (boolean):
+  - **Description:** Automatically switch between Pro and Flash models based on
+    Plan Mode status. Uses Pro for the planning phase and Flash for the
+    implementation phase.
+  - **Default:** `true`
+
 - **`general.retryFetchErrors`** (boolean):
   - **Description:** Retry on "exception TypeError: fetch failed sending
     request" errors.

packages/cli/src/config/settingsSchema.ts

@@ -285,6 +285,16 @@ const SETTINGS_SCHEMA = {
               'The directory where planning artifacts are stored. If not specified, defaults to the system temporary directory.',
             showInDialog: true,
           },
+          modelRouting: {
+            type: 'boolean',
+            label: 'Plan Model Routing',
+            category: 'General',
+            requiresRestart: false,
+            default: true,
+            description:
+              'Automatically switch between Pro and Flash models based on Plan Mode status. Uses Pro for the planning phase and Flash for the implementation phase.',
+            showInDialog: true,
+          },
         },
       },
       retryFetchErrors: {

packages/cli/src/test-utils/mockConfig.ts

@@ -47,6 +47,8 @@ export const createMockConfig = (overrides: Partial<Config> = {}): Config =>
     setRemoteAdminSettings: vi.fn(),
     isYoloModeDisabled: vi.fn(() => false),
     isPlanEnabled: vi.fn(() => false),
+    getPlanModeRoutingEnabled: vi.fn().mockResolvedValue(true),
+    getApprovedPlanPath: vi.fn(() => undefined),
     getCoreTools: vi.fn(() => []),
     getAllowedTools: vi.fn(() => []),
     getApprovalMode: vi.fn(() => 'default'),