fix(ai): support gemini-3.1 flash-lite and thinking matching

closes #1785

packages/ai/CHANGELOG.md

@@ -2,6 +2,14 @@
 
 ## [Unreleased]
 
+### Added
+
+- Added `gemini-3.1-flash-lite-preview` fallback model entry for the `google` provider so it remains selectable until upstream model catalogs include it ([#1785](https://github.com/badlogic/pi-mono/issues/1785), thanks [@n-WN](https://github.com/n-WN)).
+
+### Fixed
+
+- Fixed Gemini 3.1 thinking-level detection in `google` and `google-vertex` providers so `gemini-3.1-*` models use Gemini 3 level-based thinking config instead of budget fallback ([#1785](https://github.com/badlogic/pi-mono/issues/1785), thanks [@n-WN](https://github.com/n-WN)).
+
 ## [0.55.4] - 2026-03-02
 
 ## [0.55.3] - 2026-02-27

packages/ai/scripts/generate-models.ts

@@ -742,6 +742,27 @@ async function generateModels() {
 		});
 	}
 
+	// Add missing Gemini 3.1 Flash Lite Preview until models.dev includes it.
+	if (!allModels.some((m) => m.provider === "google" && m.id === "gemini-3.1-flash-lite-preview")) {
+		allModels.push({
+			id: "gemini-3.1-flash-lite-preview",
+			name: "Gemini 3.1 Flash Lite Preview",
+			api: "google-generative-ai",
+			baseUrl: "https://generativelanguage.googleapis.com/v1beta",
+			provider: "google",
+			reasoning: true,
+			input: ["text", "image"],
+			cost: {
+				input: 0,
+				output: 0,
+				cacheRead: 0,
+				cacheWrite: 0,
+			},
+			contextWindow: 1048576,
+			maxTokens: 65536,
+		});
+	}
+
 	// Add missing gpt models
 	if (!allModels.some(m => m.provider === "openai" && m.id === "gpt-5-chat-latest")) {
 		allModels.push({

packages/ai/src/models.generated.ts

@@ -3171,6 +3171,23 @@ export const MODELS = {
 			contextWindow: 1000000,
 			maxTokens: 64000,
 		} satisfies Model<"google-generative-ai">,
+		"gemini-3.1-flash-lite-preview": {
+			id: "gemini-3.1-flash-lite-preview",
+			name: "Gemini 3.1 Flash Lite Preview",
+			api: "google-generative-ai",
+			provider: "google",
+			baseUrl: "https://generativelanguage.googleapis.com/v1beta",
+			reasoning: true,
+			input: ["text", "image"],
+			cost: {
+				input: 0,
+				output: 0,
+				cacheRead: 0,
+				cacheWrite: 0,
+			},
+			contextWindow: 1048576,
+			maxTokens: 65536,
+		} satisfies Model<"google-generative-ai">,
 		"gemini-3.1-pro-preview": {
 			id: "gemini-3.1-pro-preview",
 			name: "Gemini 3.1 Pro Preview",
@@ -10162,13 +10179,13 @@ export const MODELS = {
 			reasoning: true,
 			input: ["text"],
 			cost: {
-				input: 0.35,
+				input: 0.39,
-				output: 1.71,
+				output: 1.9,
 				cacheRead: 0,
 				cacheWrite: 0,
 			},
-			contextWindow: 202752,
+			contextWindow: 204800,
-			maxTokens: 131072,
+			maxTokens: 204800,
 		} satisfies Model<"openai-completions">,
 		"z-ai/glm-4.6:exacto": {
 			id: "z-ai/glm-4.6:exacto",

packages/ai/src/providers/google-vertex.ts

@@ -416,11 +416,11 @@ function buildParams(
 type ClampedThinkingLevel = Exclude<PiThinkingLevel, "xhigh">;
 
 function isGemini3ProModel(model: Model<"google-generative-ai">): boolean {
-	return model.id.includes("3-pro");
+	return /gemini-3(?:\.\d+)?-pro/.test(model.id.toLowerCase());
 }
 
 function isGemini3FlashModel(model: Model<"google-generative-ai">): boolean {
-	return model.id.includes("3-flash");
+	return /gemini-3(?:\.\d+)?-flash/.test(model.id.toLowerCase());
 }
 
 function getGemini3ThinkingLevel(

packages/ai/src/providers/google.ts

@@ -386,11 +386,11 @@ function buildParams(
 type ClampedThinkingLevel = Exclude<ThinkingLevel, "xhigh">;
 
 function isGemini3ProModel(model: Model<"google-generative-ai">): boolean {
-	return model.id.includes("3-pro");
+	return /gemini-3(?:\.\d+)?-pro/.test(model.id.toLowerCase());
 }
 
 function isGemini3FlashModel(model: Model<"google-generative-ai">): boolean {
-	return model.id.includes("3-flash");
+	return /gemini-3(?:\.\d+)?-flash/.test(model.id.toLowerCase());
 }
 
 function getGemini3ThinkingLevel(