@ai-sdk/google 4.0.44 → 4.0.46

package/docs/15-google.mdx

@@ -68,11 +68,11 @@ You can use the following optional settings to customize the Google provider ins
 ## Language Models
 
 You can create models that call the [Google Generative AI API](https://ai.google.dev/api/rest) using the provider instance.
-The first argument is the model id, e.g. `gemini-2.5-flash`.
+The first argument is the model id, e.g. `gemini-3.7-flash`.
 The models support tool calls and some have multi-modal capabilities.
 
 ```ts
-const model = google('gemini-2.5-flash');
+const model = google('gemini-3.7-flash');
 ```
 
 You can use Google language models to generate text with the `generateText` function:
@@ -82,7 +82,7 @@ import { google } from '@ai-sdk/google';
 import { generateText } from 'ai';
 
 const { text } = await generateText({
-  model: google('gemini-2.5-flash'),
+  model: google('gemini-3.7-flash'),
   prompt: 'Write a vegetarian lasagna recipe for 4 people.',
 });
 ```
@@ -107,7 +107,7 @@ You can pass them as an options argument:
 ```ts
 import { google, type GoogleLanguageModelOptions } from '@ai-sdk/google';
 
-const model = google('gemini-2.5-flash');
+const model = google('gemini-3.7-flash');
 
 await generateText({
   model,
@@ -262,7 +262,7 @@ For Gemini 3 and later models, use the `thinkingLevel` parameter to control the
 import { google, GoogleLanguageModelOptions } from '@ai-sdk/google';
 import { generateText } from 'ai';
 
-const model = google('gemini-3.1-pro-preview');
+const model = google('gemini-3.7-flash');
 
 const { text, reasoning } = await generateText({
   model: model,
@@ -319,7 +319,7 @@ import { google } from '@ai-sdk/google';
 import { generateText } from 'ai';
 
 const result = await generateText({
-  model: google('gemini-2.5-flash'),
+  model: google('gemini-3.7-flash'),
   messages: [
     {
       role: 'user',
@@ -346,7 +346,7 @@ import { google } from '@ai-sdk/google';
 import { generateText } from 'ai';
 
 const result = await generateText({
-  model: google('gemini-2.5-flash'),
+  model: google('gemini-3.7-flash'),
   messages: [
     {
       role: 'user',
@@ -382,15 +382,15 @@ Google supports both explicit and implicit caching to help reduce costs on repet
 
 #### Implicit Caching
 
-Gemini 2.5 models automatically provide cache cost savings without needing to create an explicit cache. When you send requests that share common prefixes with previous requests, you'll receive a 75% token discount on cached content.
+Gemini models automatically provide cache cost savings without needing to create an explicit cache. When you send requests that share common prefixes with previous requests, you'll receive a 75% token discount on cached content.
 
 To maximize cache hits with implicit caching:
 
 - Keep content at the beginning of requests consistent
 - Add variable content (like user questions) at the end of prompts
 - Ensure requests meet minimum token requirements:
-  - Gemini 2.5 Flash: 1024 tokens minimum
-  - Gemini 2.5 Pro: 2048 tokens minimum
+  - Gemini 3 series: 4096 tokens minimum
+  - Gemini 2.5 series: 2048 tokens minimum
 
 ```ts
 import { google } from '@ai-sdk/google';
@@ -401,13 +401,13 @@ const baseContext =
   'You are a cooking assistant with expertise in Italian cuisine. Here are 1000 lasagna recipes for reference...';
 
 const { text: veggieLasagna } = await generateText({
-  model: google('gemini-2.5-pro'),
+  model: google('gemini-3.7-flash'),
   prompt: `${baseContext}\n\nWrite a vegetarian lasagna recipe for 4 people.`,
 });
 
 // Second request with same prefix - eligible for cache hit
 const { text: meatLasagna, providerMetadata } = await generateText({
-  model: google('gemini-2.5-pro'),
+  model: google('gemini-3.7-flash'),
   prompt: `${baseContext}\n\nWrite a meat lasagna recipe for 12 people.`,
 });
 
@@ -418,11 +418,11 @@ console.log('Cached tokens:', providerMetadata.google);
 //   groundingMetadata: null,
 //   safetyRatings: null,
 //   usageMetadata: {
-//     cachedContentTokenCount: 2027,
+//     cachedContentTokenCount: 5027,
 //     thoughtsTokenCount: 702,
-//     promptTokenCount: 2152,
+//     promptTokenCount: 5152,
 //     candidatesTokenCount: 710,
-//     totalTokenCount: 3564
+//     totalTokenCount: 6564
 //   }
 // }
 ```
@@ -435,7 +435,7 @@ console.log('Cached tokens:', providerMetadata.google);
 
 #### Explicit Caching
 
-For guaranteed cost savings, you can still use explicit caching with Gemini 2.5 and 2.0 models. See the [models page](https://ai.google.dev/gemini-api/docs/models) to check if caching is supported for the used model:
+For guaranteed cost savings, you can still use explicit caching with Gemini 2.5 and 3 series models. See the [models page](https://ai.google.dev/gemini-api/docs/models) to check if caching is supported for the used model:
 
 ```ts
 import { google, type GoogleLanguageModelOptions } from '@ai-sdk/google';
@@ -446,7 +446,7 @@ const ai = new GoogleGenAI({
   apiKey: process.env.GOOGLE_GENERATIVE_AI_API_KEY,
 });
 
-const model = 'gemini-2.5-pro';
+const model = 'gemini-3.7-flash';
 
 // Create a cache with the content you want to reuse
 const cache = await ai.caches.create({
@@ -495,7 +495,7 @@ import { googleTools } from '@ai-sdk/google/internal';
 import { generateText } from 'ai';
 
 const { text, toolCalls, toolResults } = await generateText({
-  model: google('gemini-2.5-pro'),
+  model: google('gemini-3.7-flash'),
   tools: { code_execution: google.tools.codeExecution({}) },
   prompt: 'Use python to calculate the 20th fibonacci number.',
 });
@@ -514,7 +514,7 @@ import { GoogleProviderMetadata } from '@ai-sdk/google';
 import { generateText } from 'ai';
 
 const { text, sources, providerMetadata } = await generateText({
-  model: google('gemini-2.5-flash'),
+  model: google('gemini-3.7-flash'),
   tools: {
     google_search: google.tools.googleSearch({}),
   },
@@ -622,7 +622,7 @@ const vertex = createGoogleVertex({
 });
 
 const { text, sources, providerMetadata } = await generateText({
-  model: vertex('gemini-2.5-flash'),
+  model: vertex('gemini-3.7-flash'),
   tools: {
     enterprise_web_search: vertex.tools.enterpriseWebSearch({}),
   },
@@ -645,7 +645,7 @@ import { google } from '@ai-sdk/google';
 import { generateText } from 'ai';
 
 const { text, sources } = await generateText({
-  model: google('gemini-2.5-pro'),
+  model: google('gemini-3.7-flash'),
   tools: {
     file_search: google.tools.fileSearch({
       fileSearchStoreNames: [
@@ -672,7 +672,7 @@ import { google } from '@ai-sdk/google';
 import { generateText } from 'ai';
 
 const { text, sources, providerMetadata } = await generateText({
-  model: google('gemini-2.5-flash'),
+  model: google('gemini-3.7-flash'),
   prompt: `Based on the document: https://ai.google.dev/gemini-api/docs/url-context.
           Answer this question: How many links we can consume in one request?`,
   tools: {
@@ -750,7 +750,7 @@ import { google } from '@ai-sdk/google';
 import { generateText } from 'ai';
 
 const { text, sources, providerMetadata } = await generateText({
-  model: google('gemini-2.5-flash'),
+  model: google('gemini-3.7-flash'),
   prompt: `Based on this context: https://ai-sdk.dev/providers/ai-sdk-providers/google, tell me how to use Gemini with AI SDK.
     Also, provide the latest news about AI SDK V5.`,
   tools: {
@@ -775,7 +775,7 @@ import { GoogleProviderMetadata } from '@ai-sdk/google';
 import { generateText } from 'ai';
 
 const { text, sources, providerMetadata } = await generateText({
-  model: google('gemini-2.5-flash'),
+  model: google('gemini-3.7-flash'),
   tools: {
     google_maps: google.tools.googleMaps({}),
   },
@@ -839,7 +839,7 @@ const vertex = createGoogleVertex({
 });
 
 const { text, sources, providerMetadata } = await generateText({
-  model: vertex('gemini-2.5-flash'),
+  model: vertex('gemini-3.7-flash'),
   tools: {
     vertex_rag_store: vertex.tools.vertexRagStore({
       ragCorpus:
@@ -1006,7 +1006,7 @@ You can disable structured outputs for object generation as a workaround:
 
 ```ts highlight="3,8"
 const { output } = await generateText({
-  model: google('gemini-2.5-flash'),
+  model: google('gemini-3.7-flash'),
   providerOptions: {
     google: {
       structuredOutputs: false,
@@ -1199,13 +1199,13 @@ import { google } from '@ai-sdk/google';
 import { generateText } from 'ai';
 
 const { text } = await generateText({
-  model: google.interactions('gemini-2.5-flash'),
+  model: google.interactions('gemini-3.7-flash'),
   prompt: 'Hello, how are you?',
 });
 ```
 
 `google.interactions(...)` accepts a model ID string (e.g.
-`'gemini-2.5-flash'`, `'gemini-3-pro-preview'`), `{ agent: <name> }` to use
+`'gemini-3.7-flash'`, `'gemini-3.1-pro-preview'`), `{ agent: <name> }` to use
 a Gemini [agent preset](#agent-presets), or `{ managedAgent: <name> }` to
 invoke a [managed agent](#managed-agents) you created on Google's side.
 The returned model can be passed to `generateText` and `streamText` like
@@ -1232,7 +1232,7 @@ import {
 import { generateText } from 'ai';
 
 await generateText({
-  model: google.interactions('gemini-2.5-flash'),
+  model: google.interactions('gemini-3.7-flash'),
   prompt: 'What color is the sky in one word?',
   providerOptions: {
     google: {
@@ -1372,7 +1372,7 @@ import {
 import { generateText } from 'ai';
 
 const turn1 = await generateText({
-  model: google.interactions('gemini-2.5-flash'),
+  model: google.interactions('gemini-3.7-flash'),
   prompt: 'What are the three largest cities in Spain?',
 });
 
@@ -1381,7 +1381,7 @@ const interactionId = turn1.providerMetadata?.google?.interactionId as
   | undefined;
 
 const turn2 = await generateText({
-  model: google.interactions('gemini-2.5-flash'),
+  model: google.interactions('gemini-3.7-flash'),
   prompt: 'What is the most famous landmark in the second one?',
   providerOptions: {
     google: {
@@ -1406,7 +1406,7 @@ const messages: Array<ModelMessage> = [
 ];
 
 const turn1 = await generateText({
-  model: google.interactions('gemini-2.5-flash'),
+  model: google.interactions('gemini-3.7-flash'),
   messages,
   providerOptions: {
     google: { store: false } satisfies GoogleLanguageModelInteractionsOptions,
@@ -1420,7 +1420,7 @@ messages.push({
 });
 
 const turn2 = await generateText({
-  model: google.interactions('gemini-2.5-flash'),
+  model: google.interactions('gemini-3.7-flash'),
   messages,
   providerOptions: {
     google: { store: false } satisfies GoogleLanguageModelInteractionsOptions,
@@ -1453,7 +1453,7 @@ import { google } from '@ai-sdk/google';
 import { generateText } from 'ai';
 
 const { text, sources } = await generateText({
-  model: google.interactions('gemini-2.5-flash'),
+  model: google.interactions('gemini-3.7-flash'),
   tools: {
     google_search: google.tools.googleSearch({}),
   },
@@ -1477,7 +1477,7 @@ const weatherTool = tool({
 });
 
 const { text, toolCalls } = await generateText({
-  model: google.interactions('gemini-2.5-flash'),
+  model: google.interactions('gemini-3.7-flash'),
   tools: { getWeather: weatherTool },
   stopWhen: stepCountIs(5),
   prompt: 'What is the weather in San Francisco right now?',
@@ -1776,7 +1776,7 @@ import { google } from '@ai-sdk/google';
 import { streamText } from 'ai';
 
 const result = streamText({
-  model: google.interactions('gemini-2.5-flash'),
+  model: google.interactions('gemini-3.7-flash'),
   prompt: 'Hello, how are you?',
 });
 

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@ai-sdk/google",
-  "version": "4.0.44",
+  "version": "4.0.46",
   "type": "module",
   "license": "Apache-2.0",
   "sideEffects": false,

package/src/google-language-model.ts

@@ -1170,7 +1170,7 @@ function resolveThinkingConfig({
     getGoogleModelCapabilities(modelId).usesGemini3Features &&
     !modelId.includes('gemini-3-pro-image')
   ) {
-    return resolveGemini3ThinkingConfig({ reasoning, warnings });
+    return resolveGemini3ThinkingConfig({ reasoning, modelId, warnings });
   }
 
   return resolveGemini25ThinkingConfig({ reasoning, modelId, warnings });
@@ -1178,23 +1178,27 @@ function resolveThinkingConfig({
 
 function resolveGemini3ThinkingConfig({
   reasoning,
+  modelId,
   warnings,
 }: {
   reasoning: Exclude<
     LanguageModelV4CallOptions['reasoning'],
     'provider-default' | undefined
   >;
+  modelId: string;
   warnings: SharedV4Warning[];
 }): Pick<GoogleThinkingConfig, 'thinkingLevel'> | undefined {
+  const minimumThinkingLevel = getMinimumThinkingLevelForGemini3Model(modelId);
+
   if (reasoning === 'none') {
     // It's not possible to fully disable thinking with Gemini 3.
-    return { thinkingLevel: 'minimal' };
+    return { thinkingLevel: minimumThinkingLevel };
   }
 
   const thinkingLevel = mapReasoningToProviderEffort({
     reasoning,
     effortMap: {
-      minimal: 'minimal',
+      minimal: minimumThinkingLevel,
       low: 'low',
       medium: 'medium',
       high: 'high',
@@ -1210,6 +1214,31 @@ function resolveGemini3ThinkingConfig({
   return { thinkingLevel };
 }
 
+function getMinimumThinkingLevelForGemini3Model(
+  modelId: string,
+): 'minimal' | 'low' {
+  const modelName = modelId.split('/').at(-1)?.toLowerCase();
+
+  if (modelName === 'gemini-flash-latest') {
+    return 'low';
+  }
+
+  const versionMatch = /^gemini-(\d+)\.(\d+)-flash(?:$|-(?!lite(?:-|$)))/.exec(
+    modelName ?? '',
+  );
+
+  if (versionMatch == null) {
+    return 'minimal';
+  }
+
+  const majorVersion = Number(versionMatch[1]);
+  const minorVersion = Number(versionMatch[2]);
+
+  return majorVersion > 3 || (majorVersion === 3 && minorVersion >= 7)
+    ? 'low'
+    : 'minimal';
+}
+
 function resolveGemini25ThinkingConfig({
   reasoning,
   modelId,