@ai-sdk/google 3.0.24 → 3.0.26

package/CHANGELOG.md

@@ -1,5 +1,17 @@
 # @ai-sdk/google
 
+## 3.0.26
+
+### Patch Changes
+
+- 4c27179: feat(google): allow using Gemini image models with `generateImage`
+
+## 3.0.25
+
+### Patch Changes
+
+- 99fbed8: feat: normalize provider specific model options type names and ensure they are exported
+
 ## 3.0.24
 
 ### Patch Changes

package/dist/index.d.mts

@@ -12,7 +12,7 @@ declare const googleErrorDataSchema: _ai_sdk_provider_utils.LazySchema<{
 type GoogleErrorData = InferSchema<typeof googleErrorDataSchema>;
 
 type GoogleGenerativeAIModelId = 'gemini-1.5-flash' | 'gemini-1.5-flash-latest' | 'gemini-1.5-flash-001' | 'gemini-1.5-flash-002' | 'gemini-1.5-flash-8b' | 'gemini-1.5-flash-8b-latest' | 'gemini-1.5-flash-8b-001' | 'gemini-1.5-pro' | 'gemini-1.5-pro-latest' | 'gemini-1.5-pro-001' | 'gemini-1.5-pro-002' | 'gemini-2.0-flash' | 'gemini-2.0-flash-001' | 'gemini-2.0-flash-live-001' | 'gemini-2.0-flash-lite' | 'gemini-2.0-pro-exp-02-05' | 'gemini-2.0-flash-thinking-exp-01-21' | 'gemini-2.0-flash-exp' | 'gemini-2.5-pro' | 'gemini-2.5-flash' | 'gemini-2.5-flash-image' | 'gemini-2.5-flash-lite' | 'gemini-2.5-flash-lite-preview-09-2025' | 'gemini-2.5-flash-preview-04-17' | 'gemini-2.5-flash-preview-09-2025' | 'gemini-3-pro-preview' | 'gemini-3-pro-image-preview' | 'gemini-3-flash-preview' | 'gemini-pro-latest' | 'gemini-flash-latest' | 'gemini-flash-lite-latest' | 'gemini-2.5-pro-exp-03-25' | 'gemini-exp-1206' | 'gemma-3-12b-it' | 'gemma-3-27b-it' | (string & {});
-declare const googleGenerativeAIProviderOptions: _ai_sdk_provider_utils.LazySchema<{
+declare const googleLanguageModelOptions: _ai_sdk_provider_utils.LazySchema<{
     responseModalities?: ("TEXT" | "IMAGE")[] | undefined;
     thinkingConfig?: {
         thinkingBudget?: number | undefined;
@@ -40,7 +40,7 @@ declare const googleGenerativeAIProviderOptions: _ai_sdk_provider_utils.LazySche
         } | undefined;
     } | undefined;
 }>;
-type GoogleGenerativeAIProviderOptions = InferSchema<typeof googleGenerativeAIProviderOptions>;
+type GoogleLanguageModelOptions = InferSchema<typeof googleLanguageModelOptions>;
 
 declare const responseSchema: _ai_sdk_provider_utils.LazySchema<{
     candidates: {
@@ -160,7 +160,7 @@ interface GoogleGenerativeAIProviderMetadata {
     safetyRatings: GoogleGenerativeAISafetyRating[] | null;
 }
 
-type GoogleGenerativeAIImageModelId = 'imagen-4.0-generate-001' | 'imagen-4.0-ultra-generate-001' | 'imagen-4.0-fast-generate-001' | (string & {});
+type GoogleGenerativeAIImageModelId = 'imagen-4.0-generate-001' | 'imagen-4.0-ultra-generate-001' | 'imagen-4.0-fast-generate-001' | 'gemini-2.5-flash-image' | 'gemini-3-pro-image-preview' | (string & {});
 interface GoogleGenerativeAIImageSettings {
     /**
      * Override the maximum number of images per call (default 4)
@@ -168,22 +168,22 @@ interface GoogleGenerativeAIImageSettings {
     maxImagesPerCall?: number;
 }
 
-declare const googleImageProviderOptionsSchema: _ai_sdk_provider_utils.LazySchema<{
+declare const googleImageModelOptionsSchema: _ai_sdk_provider_utils.LazySchema<{
     personGeneration?: "dont_allow" | "allow_adult" | "allow_all" | null | undefined;
     aspectRatio?: "1:1" | "3:4" | "4:3" | "9:16" | "16:9" | null | undefined;
 }>;
-type GoogleGenerativeAIImageProviderOptions = InferSchema<typeof googleImageProviderOptionsSchema>;
+type GoogleImageModelOptions = InferSchema<typeof googleImageModelOptionsSchema>;
 
 type GoogleGenerativeAIEmbeddingModelId = 'gemini-embedding-001' | 'text-embedding-004' | (string & {});
-declare const googleGenerativeAIEmbeddingProviderOptions: _ai_sdk_provider_utils.LazySchema<{
+declare const googleEmbeddingModelOptions: _ai_sdk_provider_utils.LazySchema<{
     outputDimensionality?: number | undefined;
     taskType?: "SEMANTIC_SIMILARITY" | "CLASSIFICATION" | "CLUSTERING" | "RETRIEVAL_DOCUMENT" | "RETRIEVAL_QUERY" | "QUESTION_ANSWERING" | "FACT_VERIFICATION" | "CODE_RETRIEVAL_QUERY" | undefined;
 }>;
-type GoogleGenerativeAIEmbeddingProviderOptions = InferSchema<typeof googleGenerativeAIEmbeddingProviderOptions>;
+type GoogleEmbeddingModelOptions = InferSchema<typeof googleEmbeddingModelOptions>;
 
 type GoogleGenerativeAIVideoModelId = 'veo-3.1-fast-generate-preview' | 'veo-3.1-generate-preview' | 'veo-3.1-generate' | 'veo-2.0-generate-001' | (string & {});
 
-type GoogleGenerativeAIVideoProviderOptions = {
+type GoogleVideoModelOptions = {
     pollIntervalMs?: number | null;
     pollTimeoutMs?: number | null;
     personGeneration?: 'dont_allow' | 'allow_adult' | 'allow_all' | null;
@@ -350,4 +350,4 @@ declare const google: GoogleGenerativeAIProvider;
 
 declare const VERSION: string;
 
-export { type GoogleErrorData, type GoogleGenerativeAIEmbeddingProviderOptions, type GoogleGenerativeAIImageProviderOptions, type GoogleGenerativeAIProvider, type GoogleGenerativeAIProviderMetadata, type GoogleGenerativeAIProviderOptions, type GoogleGenerativeAIProviderSettings, type GoogleGenerativeAIVideoModelId, type GoogleGenerativeAIVideoProviderOptions, VERSION, createGoogleGenerativeAI, google };
+export { type GoogleEmbeddingModelOptions, type GoogleErrorData, type GoogleEmbeddingModelOptions as GoogleGenerativeAIEmbeddingProviderOptions, type GoogleImageModelOptions as GoogleGenerativeAIImageProviderOptions, type GoogleGenerativeAIProvider, type GoogleGenerativeAIProviderMetadata, type GoogleLanguageModelOptions as GoogleGenerativeAIProviderOptions, type GoogleGenerativeAIProviderSettings, type GoogleGenerativeAIVideoModelId, type GoogleVideoModelOptions as GoogleGenerativeAIVideoProviderOptions, type GoogleImageModelOptions, type GoogleLanguageModelOptions, type GoogleVideoModelOptions, VERSION, createGoogleGenerativeAI, google };

package/dist/index.d.ts

@@ -12,7 +12,7 @@ declare const googleErrorDataSchema: _ai_sdk_provider_utils.LazySchema<{
 type GoogleErrorData = InferSchema<typeof googleErrorDataSchema>;
 
 type GoogleGenerativeAIModelId = 'gemini-1.5-flash' | 'gemini-1.5-flash-latest' | 'gemini-1.5-flash-001' | 'gemini-1.5-flash-002' | 'gemini-1.5-flash-8b' | 'gemini-1.5-flash-8b-latest' | 'gemini-1.5-flash-8b-001' | 'gemini-1.5-pro' | 'gemini-1.5-pro-latest' | 'gemini-1.5-pro-001' | 'gemini-1.5-pro-002' | 'gemini-2.0-flash' | 'gemini-2.0-flash-001' | 'gemini-2.0-flash-live-001' | 'gemini-2.0-flash-lite' | 'gemini-2.0-pro-exp-02-05' | 'gemini-2.0-flash-thinking-exp-01-21' | 'gemini-2.0-flash-exp' | 'gemini-2.5-pro' | 'gemini-2.5-flash' | 'gemini-2.5-flash-image' | 'gemini-2.5-flash-lite' | 'gemini-2.5-flash-lite-preview-09-2025' | 'gemini-2.5-flash-preview-04-17' | 'gemini-2.5-flash-preview-09-2025' | 'gemini-3-pro-preview' | 'gemini-3-pro-image-preview' | 'gemini-3-flash-preview' | 'gemini-pro-latest' | 'gemini-flash-latest' | 'gemini-flash-lite-latest' | 'gemini-2.5-pro-exp-03-25' | 'gemini-exp-1206' | 'gemma-3-12b-it' | 'gemma-3-27b-it' | (string & {});
-declare const googleGenerativeAIProviderOptions: _ai_sdk_provider_utils.LazySchema<{
+declare const googleLanguageModelOptions: _ai_sdk_provider_utils.LazySchema<{
     responseModalities?: ("TEXT" | "IMAGE")[] | undefined;
     thinkingConfig?: {
         thinkingBudget?: number | undefined;
@@ -40,7 +40,7 @@ declare const googleGenerativeAIProviderOptions: _ai_sdk_provider_utils.LazySche
         } | undefined;
     } | undefined;
 }>;
-type GoogleGenerativeAIProviderOptions = InferSchema<typeof googleGenerativeAIProviderOptions>;
+type GoogleLanguageModelOptions = InferSchema<typeof googleLanguageModelOptions>;
 
 declare const responseSchema: _ai_sdk_provider_utils.LazySchema<{
     candidates: {
@@ -160,7 +160,7 @@ interface GoogleGenerativeAIProviderMetadata {
     safetyRatings: GoogleGenerativeAISafetyRating[] | null;
 }
 
-type GoogleGenerativeAIImageModelId = 'imagen-4.0-generate-001' | 'imagen-4.0-ultra-generate-001' | 'imagen-4.0-fast-generate-001' | (string & {});
+type GoogleGenerativeAIImageModelId = 'imagen-4.0-generate-001' | 'imagen-4.0-ultra-generate-001' | 'imagen-4.0-fast-generate-001' | 'gemini-2.5-flash-image' | 'gemini-3-pro-image-preview' | (string & {});
 interface GoogleGenerativeAIImageSettings {
     /**
      * Override the maximum number of images per call (default 4)
@@ -168,22 +168,22 @@ interface GoogleGenerativeAIImageSettings {
     maxImagesPerCall?: number;
 }
 
-declare const googleImageProviderOptionsSchema: _ai_sdk_provider_utils.LazySchema<{
+declare const googleImageModelOptionsSchema: _ai_sdk_provider_utils.LazySchema<{
     personGeneration?: "dont_allow" | "allow_adult" | "allow_all" | null | undefined;
     aspectRatio?: "1:1" | "3:4" | "4:3" | "9:16" | "16:9" | null | undefined;
 }>;
-type GoogleGenerativeAIImageProviderOptions = InferSchema<typeof googleImageProviderOptionsSchema>;
+type GoogleImageModelOptions = InferSchema<typeof googleImageModelOptionsSchema>;
 
 type GoogleGenerativeAIEmbeddingModelId = 'gemini-embedding-001' | 'text-embedding-004' | (string & {});
-declare const googleGenerativeAIEmbeddingProviderOptions: _ai_sdk_provider_utils.LazySchema<{
+declare const googleEmbeddingModelOptions: _ai_sdk_provider_utils.LazySchema<{
     outputDimensionality?: number | undefined;
     taskType?: "SEMANTIC_SIMILARITY" | "CLASSIFICATION" | "CLUSTERING" | "RETRIEVAL_DOCUMENT" | "RETRIEVAL_QUERY" | "QUESTION_ANSWERING" | "FACT_VERIFICATION" | "CODE_RETRIEVAL_QUERY" | undefined;
 }>;
-type GoogleGenerativeAIEmbeddingProviderOptions = InferSchema<typeof googleGenerativeAIEmbeddingProviderOptions>;
+type GoogleEmbeddingModelOptions = InferSchema<typeof googleEmbeddingModelOptions>;
 
 type GoogleGenerativeAIVideoModelId = 'veo-3.1-fast-generate-preview' | 'veo-3.1-generate-preview' | 'veo-3.1-generate' | 'veo-2.0-generate-001' | (string & {});
 
-type GoogleGenerativeAIVideoProviderOptions = {
+type GoogleVideoModelOptions = {
     pollIntervalMs?: number | null;
     pollTimeoutMs?: number | null;
     personGeneration?: 'dont_allow' | 'allow_adult' | 'allow_all' | null;
@@ -350,4 +350,4 @@ declare const google: GoogleGenerativeAIProvider;
 
 declare const VERSION: string;
 
-export { type GoogleErrorData, type GoogleGenerativeAIEmbeddingProviderOptions, type GoogleGenerativeAIImageProviderOptions, type GoogleGenerativeAIProvider, type GoogleGenerativeAIProviderMetadata, type GoogleGenerativeAIProviderOptions, type GoogleGenerativeAIProviderSettings, type GoogleGenerativeAIVideoModelId, type GoogleGenerativeAIVideoProviderOptions, VERSION, createGoogleGenerativeAI, google };
+export { type GoogleEmbeddingModelOptions, type GoogleErrorData, type GoogleEmbeddingModelOptions as GoogleGenerativeAIEmbeddingProviderOptions, type GoogleImageModelOptions as GoogleGenerativeAIImageProviderOptions, type GoogleGenerativeAIProvider, type GoogleGenerativeAIProviderMetadata, type GoogleLanguageModelOptions as GoogleGenerativeAIProviderOptions, type GoogleGenerativeAIProviderSettings, type GoogleGenerativeAIVideoModelId, type GoogleVideoModelOptions as GoogleGenerativeAIVideoProviderOptions, type GoogleImageModelOptions, type GoogleLanguageModelOptions, type GoogleVideoModelOptions, VERSION, createGoogleGenerativeAI, google };

package/dist/index.js

@@ -30,7 +30,7 @@ module.exports = __toCommonJS(src_exports);
 var import_provider_utils16 = require("@ai-sdk/provider-utils");
 
 // src/version.ts
-var VERSION = true ? "3.0.24" : "0.0.0-test";
+var VERSION = true ? "3.0.26" : "0.0.0-test";
 
 // src/google-generative-ai-embedding-model.ts
 var import_provider = require("@ai-sdk/provider");
@@ -59,7 +59,7 @@ var googleFailedResponseHandler = (0, import_provider_utils.createJsonErrorRespo
 // src/google-generative-ai-embedding-options.ts
 var import_provider_utils2 = require("@ai-sdk/provider-utils");
 var import_v42 = require("zod/v4");
-var googleGenerativeAIEmbeddingProviderOptions = (0, import_provider_utils2.lazySchema)(
+var googleEmbeddingModelOptions = (0, import_provider_utils2.lazySchema)(
   () => (0, import_provider_utils2.zodSchema)(
     import_v42.z.object({
       /**
@@ -114,7 +114,7 @@ var GoogleGenerativeAIEmbeddingModel = class {
     const googleOptions = await (0, import_provider_utils3.parseProviderOptions)({
       provider: "google",
       providerOptions,
-      schema: googleGenerativeAIEmbeddingProviderOptions
+      schema: googleEmbeddingModelOptions
     });
     if (values.length > this.maxEmbeddingsPerCall) {
       throw new import_provider.TooManyEmbeddingValuesForCallError({
@@ -542,7 +542,7 @@ function getModelPath(modelId) {
 // src/google-generative-ai-options.ts
 var import_provider_utils5 = require("@ai-sdk/provider-utils");
 var import_v44 = require("zod/v4");
-var googleGenerativeAIProviderOptions = (0, import_provider_utils5.lazySchema)(
+var googleLanguageModelOptions = (0, import_provider_utils5.lazySchema)(
   () => (0, import_provider_utils5.zodSchema)(
     import_v44.z.object({
       responseModalities: import_v44.z.array(import_v44.z.enum(["TEXT", "IMAGE"])).optional(),
@@ -924,13 +924,13 @@ var GoogleGenerativeAILanguageModel = class {
     let googleOptions = await (0, import_provider_utils6.parseProviderOptions)({
       provider: providerOptionsName,
       providerOptions,
-      schema: googleGenerativeAIProviderOptions
+      schema: googleLanguageModelOptions
     });
     if (googleOptions == null && providerOptionsName !== "google") {
       googleOptions = await (0, import_provider_utils6.parseProviderOptions)({
         provider: "google",
         providerOptions,
-        schema: googleGenerativeAIProviderOptions
+        schema: googleLanguageModelOptions
       });
     }
     if ((tools == null ? void 0 : tools.some(
@@ -1786,13 +1786,24 @@ var GoogleGenerativeAIImageModel = class {
     this.specificationVersion = "v3";
   }
   get maxImagesPerCall() {
-    var _a;
-    return (_a = this.settings.maxImagesPerCall) != null ? _a : 4;
+    if (this.settings.maxImagesPerCall != null) {
+      return this.settings.maxImagesPerCall;
+    }
+    if (isGeminiModel(this.modelId)) {
+      return 10;
+    }
+    return 4;
   }
   get provider() {
     return this.config.provider;
   }
   async doGenerate(options) {
+    if (isGeminiModel(this.modelId)) {
+      return this.doGenerateGemini(options);
+    }
+    return this.doGenerateImagen(options);
+  }
+  async doGenerateImagen(options) {
     var _a, _b, _c;
     const {
       prompt,
@@ -1809,7 +1820,7 @@ var GoogleGenerativeAIImageModel = class {
     const warnings = [];
     if (files != null && files.length > 0) {
       throw new Error(
-        "Google Generative AI does not support image editing. Use Google Vertex AI (@ai-sdk/google-vertex) for image editing capabilities."
+        "Google Generative AI does not support image editing with Imagen models. Use Google Vertex AI (@ai-sdk/google-vertex) for image editing capabilities."
       );
     }
     if (mask != null) {
@@ -1834,7 +1845,7 @@ var GoogleGenerativeAIImageModel = class {
     const googleOptions = await (0, import_provider_utils14.parseProviderOptions)({
       provider: "google",
       providerOptions,
-      schema: googleImageProviderOptionsSchema
+      schema: googleImageModelOptionsSchema
     });
     const currentDate = (_c = (_b = (_a = this.config._internal) == null ? void 0 : _a.currentDate) == null ? void 0 : _b.call(_a)) != null ? _c : /* @__PURE__ */ new Date();
     const parameters = {
@@ -1865,10 +1876,10 @@ var GoogleGenerativeAIImageModel = class {
       images: response.predictions.map(
         (p) => p.bytesBase64Encoded
       ),
-      warnings: warnings != null ? warnings : [],
+      warnings,
       providerMetadata: {
         google: {
-          images: response.predictions.map((prediction) => ({
+          images: response.predictions.map(() => ({
             // Add any prediction-specific metadata here
           }))
         }
@@ -1880,7 +1891,113 @@ var GoogleGenerativeAIImageModel = class {
       }
     };
   }
+  async doGenerateGemini(options) {
+    var _a, _b, _c, _d, _e, _f, _g, _h, _i;
+    const {
+      prompt,
+      n,
+      size,
+      aspectRatio,
+      seed,
+      providerOptions,
+      headers,
+      abortSignal,
+      files,
+      mask
+    } = options;
+    const warnings = [];
+    if (mask != null) {
+      throw new Error(
+        "Gemini image models do not support mask-based image editing."
+      );
+    }
+    if (n != null && n > 1) {
+      throw new Error(
+        "Gemini image models do not support generating a set number of images per call. Use n=1 or omit the n parameter."
+      );
+    }
+    if (size != null) {
+      warnings.push({
+        type: "unsupported",
+        feature: "size",
+        details: "This model does not support the `size` option. Use `aspectRatio` instead."
+      });
+    }
+    const userContent = [];
+    if (prompt != null) {
+      userContent.push({ type: "text", text: prompt });
+    }
+    if (files != null && files.length > 0) {
+      for (const file of files) {
+        if (file.type === "url") {
+          userContent.push({
+            type: "file",
+            data: new URL(file.url),
+            mediaType: "image/*"
+          });
+        } else {
+          userContent.push({
+            type: "file",
+            data: typeof file.data === "string" ? file.data : new Uint8Array(file.data),
+            mediaType: file.mediaType
+          });
+        }
+      }
+    }
+    const languageModelPrompt = [
+      { role: "user", content: userContent }
+    ];
+    const languageModel = new GoogleGenerativeAILanguageModel(this.modelId, {
+      provider: this.config.provider,
+      baseURL: this.config.baseURL,
+      headers: (_a = this.config.headers) != null ? _a : {},
+      fetch: this.config.fetch,
+      generateId: (_b = this.config.generateId) != null ? _b : import_provider_utils14.generateId
+    });
+    const result = await languageModel.doGenerate({
+      prompt: languageModelPrompt,
+      seed,
+      providerOptions: {
+        google: {
+          responseModalities: ["IMAGE"],
+          imageConfig: aspectRatio ? { aspectRatio } : void 0,
+          ...(_c = providerOptions == null ? void 0 : providerOptions.google) != null ? _c : {}
+        }
+      },
+      headers,
+      abortSignal
+    });
+    const currentDate = (_f = (_e = (_d = this.config._internal) == null ? void 0 : _d.currentDate) == null ? void 0 : _e.call(_d)) != null ? _f : /* @__PURE__ */ new Date();
+    const images = [];
+    for (const part of result.content) {
+      if (part.type === "file" && part.mediaType.startsWith("image/")) {
+        images.push((0, import_provider_utils14.convertToBase64)(part.data));
+      }
+    }
+    return {
+      images,
+      warnings,
+      providerMetadata: {
+        google: {
+          images: images.map(() => ({}))
+        }
+      },
+      response: {
+        timestamp: currentDate,
+        modelId: this.modelId,
+        headers: (_g = result.response) == null ? void 0 : _g.headers
+      },
+      usage: result.usage ? {
+        inputTokens: result.usage.inputTokens.total,
+        outputTokens: result.usage.outputTokens.total,
+        totalTokens: ((_h = result.usage.inputTokens.total) != null ? _h : 0) + ((_i = result.usage.outputTokens.total) != null ? _i : 0)
+      } : void 0
+    };
+  }
 };
+function isGeminiModel(modelId) {
+  return modelId.startsWith("gemini-");
+}
 var googleImageResponseSchema = (0, import_provider_utils14.lazySchema)(
   () => (0, import_provider_utils14.zodSchema)(
     import_v413.z.object({
@@ -1888,7 +2005,7 @@ var googleImageResponseSchema = (0, import_provider_utils14.lazySchema)(
     })
   )
 );
-var googleImageProviderOptionsSchema = (0, import_provider_utils14.lazySchema)(
+var googleImageModelOptionsSchema = (0, import_provider_utils14.lazySchema)(
   () => (0, import_provider_utils14.zodSchema)(
     import_v413.z.object({
       personGeneration: import_v413.z.enum(["dont_allow", "allow_adult", "allow_all"]).nullish(),
@@ -1920,7 +2037,7 @@ var GoogleGenerativeAIVideoModel = class {
     const googleOptions = await (0, import_provider_utils15.parseProviderOptions)({
       provider: "google",
       providerOptions: options.providerOptions,
-      schema: googleVideoProviderOptionsSchema
+      schema: googleVideoModelOptionsSchema
     });
     const instances = [{}];
     const instance = instances[0];
@@ -2132,7 +2249,7 @@ var googleOperationSchema = import_v414.z.object({
     }).nullish()
   }).nullish()
 });
-var googleVideoProviderOptionsSchema = (0, import_provider_utils15.lazySchema)(
+var googleVideoModelOptionsSchema = (0, import_provider_utils15.lazySchema)(
   () => (0, import_provider_utils15.zodSchema)(
     import_v414.z.object({
       pollIntervalMs: import_v414.z.number().positive().nullish(),