fix(ai): prevent removed xAI models regenerating, closes #6736

packages/ai/CHANGELOG.md

@@ -5,6 +5,7 @@
 ### Fixed
 
 - Fixed Kimi Coding K3 thinking-level metadata to expose only the supported `max` level ([#6737](https://github.com/earendil-works/pi/issues/6737)).
+- Fixed catalog generation restoring xAI models removed in 0.80.9 ([#6736](https://github.com/earendil-works/pi/issues/6736)).
 
 ## [0.80.9] - 2026-07-16
 

packages/ai/README.md

@@ -747,7 +747,7 @@ Many models support thinking/reasoning capabilities where they can show their in
 const model = models.getModel('anthropic', 'claude-sonnet-4-5')!;
 // or models.getModel('openai', 'gpt-5-mini');
 // or models.getModel('google', 'gemini-2.5-flash');
-// or models.getModel('xai', 'grok-code-fast-1');
+// or models.getModel('xai', 'grok-4.5');
 
 // Check if model supports reasoning
 if (model.reasoning) {

packages/ai/scripts/generate-models.ts

@@ -302,6 +302,13 @@ const OPENAI_RESPONSES_NONE_REASONING_MODELS = new Set([
 	"gpt-5.6-luna",
 ]);
 const XAI_RESPONSES_MODEL_ID = "grok-4.5";
+const XAI_BUILTIN_EXCLUDED_MODEL_IDS = new Set([
+	"grok-3",
+	"grok-3-fast",
+	"grok-4.20-0309-non-reasoning",
+	"grok-4.20-0309-reasoning",
+	"grok-code-fast-1",
+]);
 const XAI_RESPONSES_EFFORT_LEVEL_MAP = {
 	off: null,
 	minimal: null,
@@ -1783,6 +1790,7 @@ async function generateModels() {
 	// Combine models (models.dev has priority)
 	const allModels = [...modelsDevModels, ...openRouterModels, ...aiGatewayModels].filter(
 		(model) =>
+			!(model.provider === "xai" && XAI_BUILTIN_EXCLUDED_MODEL_IDS.has(model.id)) &&
 			!((model.provider === "opencode" || model.provider === "opencode-go") && model.id === "gpt-5.3-codex-spark"),
 	);
 
@@ -2142,56 +2150,6 @@ async function generateModels() {
 	];
 	allModels.push(...codexModels);
 
-	// Add missing Grok models
-	const missingGrokModels: Model<"openai-completions">[] = [
-		{
-			id: "grok-3",
-			name: "Grok 3",
-			api: "openai-completions",
-			baseUrl: "https://api.x.ai/v1",
-			provider: "xai",
-			reasoning: false,
-			input: ["text"],
-			cost: { input: 3, output: 15, cacheRead: 0.75, cacheWrite: 0 },
-			contextWindow: 131072,
-			maxTokens: 8192,
-		},
-		{
-			id: "grok-3-fast",
-			name: "Grok 3 Fast",
-			api: "openai-completions",
-			baseUrl: "https://api.x.ai/v1",
-			provider: "xai",
-			reasoning: false,
-			input: ["text"],
-			cost: { input: 5, output: 25, cacheRead: 1.25, cacheWrite: 0 },
-			contextWindow: 131072,
-			maxTokens: 8192,
-		},
-		{
-			id: "grok-code-fast-1",
-			name: "Grok Code Fast 1",
-			api: "openai-completions",
-			baseUrl: "https://api.x.ai/v1",
-			provider: "xai",
-			reasoning: false,
-			input: ["text"],
-			cost: {
-				input: 0.2,
-				output: 1.5,
-				cacheRead: 0.02,
-				cacheWrite: 0,
-			},
-			contextWindow: 32768,
-			maxTokens: 8192,
-		},
-	];
-	for (const model of missingGrokModels) {
-		if (!allModels.some(m => m.provider === model.provider && m.id === model.id)) {
-			allModels.push(model);
-		}
-	}
-
 	// Add missing Mistral Medium 3.5 model until models.dev includes it
 	if (!allModels.some(m => m.provider === "mistral" && m.id === "mistral-medium-3.5")) {
 		allModels.push({

packages/ai/src/providers/openrouter.models.ts

@@ -4706,9 +4706,9 @@ export const OPENROUTER_MODELS = {
 		thinkingLevelMap: {"xhigh":"xhigh"},
 		input: ["text"],
 		cost: {
-			input: 0.9212,
+			input: 0.9198,
-			output: 2.8952,
+			output: 2.8908,
-			cacheRead: 0.17108,
+			cacheRead: 0.17082,
 			cacheWrite: 0,
 		},
 		contextWindow: 1048576,

packages/ai/src/providers/xai.models.ts

@@ -4,78 +4,6 @@
 import type { Model } from "../types.ts";
 
 export const XAI_MODELS = {
-	"grok-3": {
-		id: "grok-3",
-		name: "Grok 3",
-		api: "openai-completions",
-		provider: "xai",
-		baseUrl: "https://api.x.ai/v1",
-		compat: {"supportsStore":false,"supportsDeveloperRole":false,"supportsReasoningEffort":false},
-		reasoning: false,
-		input: ["text"],
-		cost: {
-			input: 3,
-			output: 15,
-			cacheRead: 0.75,
-			cacheWrite: 0,
-		},
-		contextWindow: 131072,
-		maxTokens: 8192,
-	} satisfies Model<"openai-completions">,
-	"grok-3-fast": {
-		id: "grok-3-fast",
-		name: "Grok 3 Fast",
-		api: "openai-completions",
-		provider: "xai",
-		baseUrl: "https://api.x.ai/v1",
-		compat: {"supportsStore":false,"supportsDeveloperRole":false,"supportsReasoningEffort":false},
-		reasoning: false,
-		input: ["text"],
-		cost: {
-			input: 5,
-			output: 25,
-			cacheRead: 1.25,
-			cacheWrite: 0,
-		},
-		contextWindow: 131072,
-		maxTokens: 8192,
-	} satisfies Model<"openai-completions">,
-	"grok-4.20-0309-non-reasoning": {
-		id: "grok-4.20-0309-non-reasoning",
-		name: "Grok 4.20 (Non-Reasoning)",
-		api: "openai-completions",
-		provider: "xai",
-		baseUrl: "https://api.x.ai/v1",
-		compat: {"supportsStore":false,"supportsDeveloperRole":false,"supportsReasoningEffort":false},
-		reasoning: false,
-		input: ["text", "image"],
-		cost: {
-			input: 1.25,
-			output: 2.5,
-			cacheRead: 0.2,
-			cacheWrite: 0,
-		},
-		contextWindow: 1000000,
-		maxTokens: 30000,
-	} satisfies Model<"openai-completions">,
-	"grok-4.20-0309-reasoning": {
-		id: "grok-4.20-0309-reasoning",
-		name: "Grok 4.20 (Reasoning)",
-		api: "openai-completions",
-		provider: "xai",
-		baseUrl: "https://api.x.ai/v1",
-		compat: {"supportsStore":false,"supportsDeveloperRole":false,"supportsReasoningEffort":false},
-		reasoning: true,
-		input: ["text", "image"],
-		cost: {
-			input: 1.25,
-			output: 2.5,
-			cacheRead: 0.2,
-			cacheWrite: 0,
-		},
-		contextWindow: 1000000,
-		maxTokens: 30000,
-	} satisfies Model<"openai-completions">,
 	"grok-4.3": {
 		id: "grok-4.3",
 		name: "Grok 4.3",
@@ -131,22 +59,4 @@ export const XAI_MODELS = {
 		contextWindow: 256000,
 		maxTokens: 256000,
 	} satisfies Model<"openai-completions">,
-	"grok-code-fast-1": {
-		id: "grok-code-fast-1",
-		name: "Grok Code Fast 1",
-		api: "openai-completions",
-		provider: "xai",
-		baseUrl: "https://api.x.ai/v1",
-		compat: {"supportsStore":false,"supportsDeveloperRole":false,"supportsReasoningEffort":false},
-		reasoning: false,
-		input: ["text"],
-		cost: {
-			input: 0.2,
-			output: 1.5,
-			cacheRead: 0.02,
-			cacheWrite: 0,
-		},
-		contextWindow: 32768,
-		maxTokens: 8192,
-	} satisfies Model<"openai-completions">,
 } as const;

packages/ai/test/context-overflow.test.ts

@@ -262,7 +262,7 @@ describe("Context overflow error handling", () => {
 	// =============================================================================
 
 	describe.skipIf(!process.env.XAI_API_KEY)("xAI", () => {
-		it("grok-3-fast - should detect overflow via isContextOverflow", async () => {
+		it("grok-4.3 - should detect overflow via isContextOverflow", async () => {
 			const model = getModel("xai", "grok-4.3");
 			const result = await testContextOverflow(model, process.env.XAI_API_KEY!);
 			logResult(result);

packages/ai/test/cross-provider-handoff.test.ts

@@ -79,7 +79,7 @@ const PROVIDER_MODEL_PAIRS: ProviderModelPair[] = [
 		label: "bedrock-claude-sonnet-4-5",
 	},
 	// xAI
-	{ provider: "xai", model: "grok-code-fast-1", label: "xai-grok-code-fast-1" },
+	{ provider: "xai", model: "grok-4.3", label: "xai-grok-4.3" },
 	// Cerebras
 	{ provider: "cerebras", model: "zai-glm-4.7", label: "cerebras-zai-glm-4.7" },
 	// Cloudflare Workers AI

packages/ai/test/stream.test.ts

@@ -542,7 +542,7 @@ describe("Generate E2E Tests", () => {
 		});
 	});
 
-	describe.skipIf(!process.env.XAI_API_KEY)("xAI Provider (grok-code-fast-1 via OpenAI Completions)", () => {
+	describe.skipIf(!process.env.XAI_API_KEY)("xAI Provider (grok-4.3 via OpenAI Completions)", () => {
 		const llm = getModel("xai", "grok-4.3");
 
 		it("should complete basic text generation", { retry: 3 }, async () => {

packages/ai/test/total-tokens.test.ts

@@ -244,22 +244,18 @@ describe("totalTokens field", () => {
 	// =========================================================================
 
 	describe.skipIf(!process.env.XAI_API_KEY)("xAI", () => {
-		it(
+		it("grok-4.3 - should return totalTokens equal to sum of components", { retry: 3, timeout: 60000 }, async () => {
-			"grok-3-fast - should return totalTokens equal to sum of components",
+			const llm = getModel("xai", "grok-4.3");
-			{ retry: 3, timeout: 60000 },
-			async () => {
-				const llm = getModel("xai", "grok-4.3");
 
-				console.log(`\nxAI / ${llm.id}:`);
+			console.log(`\nxAI / ${llm.id}:`);
-				const { first, second } = await testTotalTokensWithCache(llm, { apiKey: process.env.XAI_API_KEY });
+			const { first, second } = await testTotalTokensWithCache(llm, { apiKey: process.env.XAI_API_KEY });
 
-				logUsage("First request", first);
+			logUsage("First request", first);
-				logUsage("Second request", second);
+			logUsage("Second request", second);
 
-				assertTotalTokensEqualsComponents(first);
+			assertTotalTokensEqualsComponents(first);
-				assertTotalTokensEqualsComponents(second);
+			assertTotalTokensEqualsComponents(second);
-			},
+		});
-		);
 	});
 
 	// =========================================================================

packages/ai/test/xai-responses.test.ts

@@ -60,6 +60,18 @@ describe("xAI Responses provider", () => {
 		vi.restoreAllMocks();
 	});
 
+	it("excludes retired and redundant models from the built-in catalog", () => {
+		for (const modelId of [
+			"grok-3",
+			"grok-3-fast",
+			"grok-4.20-0309-non-reasoning",
+			"grok-4.20-0309-reasoning",
+			"grok-code-fast-1",
+		]) {
+			expect(Object.keys(XAI_MODELS)).not.toContain(modelId);
+		}
+	});
+
 	it("uses Responses with low/medium/high efforts only for Grok 4.5", () => {
 		expect(XAI_MODELS["grok-4.5"].api).toBe("openai-responses");
 		expect(getSupportedThinkingLevels(XAI_MODELS["grok-4.5"])).toEqual(["low", "medium", "high"]);

packages/coding-agent/CHANGELOG.md

@@ -2,6 +2,10 @@
 
 ## [Unreleased]
 
+### Fixed
+
+- Fixed inherited catalog generation restoring xAI models removed in 0.80.9 ([#6736](https://github.com/earendil-works/pi/issues/6736)).
+
 ## [0.80.9] - 2026-07-16
 
 ### New Features