replay

Remove ‘-beta’ from grok model id (#4124 )
* Remove ‘-beta’ from grok model id * Create purple-islands-move.md
2026-09-04 11:44:01 +08:00 · 2025-06-09 22:56:53 -07:00 · 2025-06-09 18:31:01 -07:00 · 2025-06-09 16:31:39 -07:00
25 changed files with 84 additions and 79 deletions
@@ -1,5 +0,0 @@
---
-"claude-dev": patch
---
-
-Migrated updateSettings to protos, removed didUpdateSettings, altered Plan/Act toggling in settings menu
@@ -1,5 +0,0 @@
---
-"claude-dev": patch
---
-
-Fixed issue where telemetry warning popup was created for every new Cline window
@@ -1,5 +0,0 @@
---
-"claude-dev": minor
---
-
-Prioritize active files in file context menu
@@ -1,5 +0,0 @@
---
-"claude-dev": patch
---
-
-toggleWorkflow protobus migration
@@ -1,5 +0,0 @@
---
-"claude-dev": patch
---
-
-Added promise to task init to prevent race condition with checkpoints
@@ -1,5 +0,0 @@
---
-"claude-dev": patch
---
-
-Spring cleaning
@@ -1,5 +0,0 @@
---
-"claude-dev": patch
---
-
-Add free grok model
@@ -0,0 +1,5 @@
+---
+"claude-dev": patch
+---
+
+Remove ‘-beta’ from grok model id
@@ -1,5 +0,0 @@
---
-"claude-dev": patch
---
-
-Fix bug where replace_in_file would not be able to handle for out-of-order SEARCH/REPLACE blocks
@@ -1,5 +0,0 @@
---
-"claude-dev": minor
---
-
-Context menu is default to File option on start up
@@ -1,5 +0,0 @@
---
-"claude-dev": minor
---
-
-Migrate fetchLatestServersFromHub to protobus
@@ -1,5 +0,0 @@
---
-"claude-dev": minor
---
-
-the response of the mcps is displayed with a collapsible which allows to focus on the model responses.
@@ -1,5 +1,14 @@
 # Changelog

+## [3.17.12]
+
+-   **Free Grok Model Available!** Access Grok 3 completely free through the Cline provider
+-   Add collapsible MCP response panels to keep conversations focused on the main AI responses while still allowing access to detailed MCP output (Thanks @valinha!)
+-   Prioritize active files (open tabs) at the top of the file context menu when using @ mentions (Thanks @abeatrix!)
+-   Fix context menu to properly default to "File" option instead of incorrectly selecting "Git Commits"
+-   Fix diff editing to handle out-of-order SEARCH/REPLACE blocks, improving reliability with models that don't follow strict ordering
+-   Fix telemetry warning popup appearing repeatedly for users who have telemetry disabled
+
 ## [3.17.11]

 -   Add support for Gemini 2.5 Pro Preview 06-05 model to Vertex AI and Google Gemini providers
@@ -13,6 +13,7 @@ interface RunDiffEvalOptions {
 	verbose: boolean
 	testPath: string
 	outputPath: string
+	replay: boolean
 }

 export async function runDiffEvalHandler(options: RunDiffEvalOptions) {
@@ -50,6 +51,10 @@ export async function runDiffEvalHandler(options: RunDiffEvalOptions) {
 		args.push("--parallel")
 	}

+	if (options.replay) {
+		args.push("--replay")
+	}
+
 	if (options.verbose) {
 		args.push("--verbose")
 	}
@@ -91,6 +91,7 @@ program
 	.option("--diff-edit-function <name>", "The diff editing function to use", "constructNewFileContentV2")
 	.option("--thinking-budget <tokens>", "Set the thinking tokens budget", "0")
 	.option("--parallel", "Run tests in parallel", false)
+	.option("--replay", "Run evaluation from a pre-recorded LLM output, skipping the API call", false)
 	.option("-v, --verbose", "Enable verbose logging", false)
 	.action(async (options) => {
 		try {
@@ -114,10 +114,10 @@ export async function runSingleEvaluation(input: TestInput): Promise<TestResult>
 			parsingFunction,
 			diffEditFunction,
 			thinkingBudgetTokens,
+			originalDiffEditToolCallMessage,
 		} = input

 		const requiredParams = {
-			apiKey,
 			systemPrompt,
 			messages,
 			modelId,
@@ -163,17 +163,26 @@ export async function runSingleEvaluation(input: TestInput): Promise<TestResult>
 			},
 		}

-		const openRouterHandler = new OpenRouterHandler(options)
-
 		// Get the output of streaming output of this llm call
 		let streamResult: StreamResult
-		try {
-			streamResult = await processStream(openRouterHandler, systemPrompt, messages)
-		} catch (error: any) {
-			return {
-				success: false,
-				error: "llm_stream_error",
-				errorString: error.message || error.toString(),
+		if (originalDiffEditToolCallMessage !== undefined) {
+			// Replay mode: mock the stream result
+			streamResult = {
+				assistantMessage: originalDiffEditToolCallMessage,
+				reasoningMessage: "",
+				usage: { inputTokens: 0, outputTokens: 0, cacheWriteTokens: 0, cacheReadTokens: 0, totalCost: 0 },
+			}
+		} else {
+			// Live mode: existing API call logic
+			try {
+				const openRouterHandler = new OpenRouterHandler(options)
+				streamResult = await processStream(openRouterHandler, systemPrompt, messages)
+			} catch (error: any) {
+				return {
+					success: false,
+					error: "llm_stream_error",
+					errorString: error.message || error.toString(),
+				}
 			}
 		}

@@ -22,12 +22,14 @@ const systemPromptGeneratorLookup: Record<string, ConstructSystemPromptFn> = {
 type TestResultSet = { [test_id: string]: (TestResult & { test_id?: string })[] }

 class NodeTestRunner {
-	private apiKey: string
+	private apiKey: string | undefined

-	constructor() {
-		this.apiKey = process.env.OPENROUTER_API_KEY!
-		if (!this.apiKey) {
-			throw new Error("OPENROUTER_API_KEY environment variable not set")
+	constructor(isReplay: boolean) {
+		if (!isReplay) {
+			this.apiKey = process.env.OPENROUTER_API_KEY
+			if (!this.apiKey) {
+				throw new Error("OPENROUTER_API_KEY environment variable not set for a non-replay run.")
+			}
 		}
 	}

@@ -125,6 +127,14 @@ class NodeTestRunner {
 	 * Run a single test example
 	 */
 	async runSingleTest(testCase: ProcessedTestCase, testConfig: TestConfig): Promise<TestResult> {
+		if (testConfig.replay && !testCase.original_diff_edit_tool_call_message) {
+			return {
+				success: false,
+				error: "missing_original_diff_edit_tool_call_message",
+				errorString: `Test case ${testCase.test_id} is missing 'original_diff_edit_tool_call_message' for replay.`,
+			}
+		}
+
 		const customSystemPrompt = this.constructSystemPrompt(testCase.system_prompt_details, testConfig.system_prompt_name)

 		// messages don't include system prompt and are everything up to the first replace_in_file tool call which results in a diff edit error
@@ -138,6 +148,7 @@ class NodeTestRunner {
 			parsingFunction: testConfig.parsing_function,
 			diffEditFunction: testConfig.diff_edit_function,
 			thinkingBudgetTokens: testConfig.thinking_tokens_budget,
+			originalDiffEditToolCallMessage: testConfig.replay ? testCase.original_diff_edit_tool_call_message : undefined,
 		}

 		return await runSingleEvaluation(input)
@@ -320,6 +331,7 @@ async function main() {
 		.option("--diff-edit-function <name>", "The diff editing function to use", "constructNewFileContentV2")
 		.option("--thinking-budget <tokens>", "Set the thinking tokens budget", "0")
 		.option("--parallel", "Run tests in parallel", false)
+		.option("--replay", "Run evaluation from a pre-recorded LLM output, skipping the API call", false)
 		.option("-v, --verbose", "Enable verbose logging", false)

 	program.parse(process.argv)
@@ -336,12 +348,13 @@ async function main() {
 		parsing_function: options.parsingFunction,
 		diff_edit_function: options.diffEditFunction,
 		thinking_tokens_budget: parseInt(options.thinkingBudget, 10),
+		replay: options.replay,
 	}

 	try {
 		const startTime = Date.now()

-		const runner = new NodeTestRunner()
+		const runner = new NodeTestRunner(testConfig.replay)
 		const testCases = runner.loadTestCases(testPath)

 		const processedTestCases: ProcessedTestCase[] = testCases.map((tc) => ({
@@ -351,6 +364,9 @@ async function main() {

 		log(isVerbose, `-Loaded ${testCases.length} test cases.`)
 		log(isVerbose, `-Executing ${testConfig.number_of_runs} run(s) per test case.`)
+		if (testConfig.replay) {
+			log(isVerbose, `-Running in REPLAY mode. No API calls will be made.`)
+		}
 		log(isVerbose, "Starting tests...\n")

 		const results = options.parallel
@@ -13,6 +13,7 @@ export interface ProcessedTestCase {
 	file_contents: string
 	file_path: string
 	system_prompt_details: SystemPromptDetails
+	original_diff_edit_tool_call_message: string
 }

 export interface TestCase {
@@ -21,6 +22,7 @@ export interface TestCase {
 	file_contents: string
 	file_path: string
 	system_prompt_details: SystemPromptDetails
+	original_diff_edit_tool_call_message: string
 }

 export interface TestConfig {
@@ -30,6 +32,7 @@ export interface TestConfig {
 	parsing_function: string
 	diff_edit_function: string
 	thinking_tokens_budget: number
+	replay: boolean
 }

 export interface SystemPromptDetails {
@@ -72,7 +75,7 @@ export interface ExtractedToolCall {
 }

 export interface TestInput {
-	apiKey: string
+	apiKey?: string
 	systemPrompt: string
 	messages: Anthropic.Messages.MessageParam[]
 	modelId: string
@@ -81,4 +84,5 @@ export interface TestInput {
 	parsingFunction: string
 	diffEditFunction: string
 	thinkingBudgetTokens: number
+	originalDiffEditToolCallMessage?: string
 }
@@ -1,12 +1,12 @@
 {
 	"name": "claude-dev",
-	"version": "3.17.11",
+	"version": "3.17.12",
 	"lockfileVersion": 2,
 	"requires": true,
 	"packages": {
 		"": {
 			"name": "claude-dev",
-			"version": "3.17.11",
+			"version": "3.17.12",
 			"license": "Apache-2.0",
 			"dependencies": {
 				"@anthropic-ai/bedrock-sdk": "^0.12.4",
@@ -2,7 +2,7 @@
 	"name": "claude-dev",
 	"displayName": "Cline",
 	"description": "Autonomous coding agent right in your IDE, capable of creating/editing files, running commands, using the browser, and more with your permission every step of the way.",
-	"version": "3.17.11",
+	"version": "3.17.12",
 	"icon": "assets/icons/icon.png",
 	"engines": {
 		"vscode": "^1.84.0"
@@ -134,7 +134,10 @@ export class ClineHandler implements ApiHandler {
 	}

 	getModel(): { id: string; info: ModelInfo } {
-		const modelId = this.options.openRouterModelId
+		let modelId = this.options.openRouterModelId
+		if (modelId === "x-ai/grok-3") {
+			modelId = "x-ai/grok-3-beta"
+		}
 		const modelInfo = this.options.openRouterModelInfo
 		if (modelId && modelInfo) {
 			return { id: modelId, info: modelInfo }
@@ -139,7 +139,10 @@ export class OpenRouterHandler implements ApiHandler {
 	}

 	getModel(): { id: string; info: ModelInfo } {
-		const modelId = this.options.openRouterModelId
+		let modelId = this.options.openRouterModelId
+		if (modelId === "x-ai/grok-3") {
+			modelId = "x-ai/grok-3-beta"
+		}
 		const modelInfo = this.options.openRouterModelInfo
 		if (modelId && modelInfo) {
 			return { id: modelId, info: modelInfo }
@@ -117,6 +117,11 @@ export async function refreshOpenRouterModels(
 						break
 				}

+				// add new model id
+				if (rawModel.id === "x-ai/grok-3-beta") {
+					models["x-ai/grok-3"] = modelInfo
+				}
+
 				models[rawModel.id] = modelInfo
 			}
 		} else {
@@ -2577,7 +2577,7 @@ export function normalizeApiConfiguration(apiConfiguration?: ApiConfiguration):
 				// TODO: remove this once we have a better way to handle free models on Cline
 				// Free grok 3 promotion
 				selectedModelInfo:
-					openRouterModelId === "x-ai/grok-3-beta"
+					openRouterModelId === "x-ai/grok-3"
 						? { ...openRouterModelInfo, inputPrice: 0, outputPrice: 0 }
 						: openRouterModelInfo,
 			}
@@ -52,7 +52,7 @@ const featuredModels = [
 		label: "Trending",
 	},
 	{
-		id: "x-ai/grok-3-beta",
+		id: "x-ai/grok-3",
 		description: "Latest flagship model from xAI, free for now!",
 		label: "Free",
 	},