@ai-sdk/google-vertex 5.0.0-beta.10 → 5.0.0-beta.108

package/src/google-vertex-provider-base.ts

@@ -0,0 +1,311 @@
+import {
+  GoogleLanguageModel,
+  GoogleSpeechModel,
+} from '@ai-sdk/google/internal';
+import type {
+  Experimental_VideoModelV4,
+  ImageModelV4,
+  LanguageModelV4,
+  ProviderV4,
+  SpeechModelV4,
+  TranscriptionModelV4,
+} from '@ai-sdk/provider';
+import {
+  generateId,
+  loadOptionalSetting,
+  loadSetting,
+  normalizeHeaders,
+  resolve,
+  withoutTrailingSlash,
+  withUserAgentSuffix,
+  type FetchFunction,
+  type Resolvable,
+} from '@ai-sdk/provider-utils';
+import { VERSION } from './version';
+import type { GoogleVertexConfig } from './google-vertex-config';
+import { GoogleVertexEmbeddingModel } from './google-vertex-embedding-model';
+import type { GoogleVertexEmbeddingModelId } from './google-vertex-embedding-model-options';
+import { GoogleVertexImageModel } from './google-vertex-image-model';
+import type { GoogleVertexImageModelId } from './google-vertex-image-settings';
+import type { GoogleVertexModelId } from './google-vertex-options';
+import { googleVertexTools } from './google-vertex-tools';
+import { GoogleVertexTranscriptionModel } from './google-vertex-transcription-model';
+import type { GoogleVertexTranscriptionModelId } from './google-vertex-transcription-model-options';
+import { GoogleVertexVideoModel } from './google-vertex-video-model';
+import type { GoogleVertexVideoModelId } from './google-vertex-video-settings';
+import type { GoogleVertexSpeechModelId } from './google-vertex-speech-model-options';
+
+const EXPRESS_MODE_BASE_URL =
+  'https://aiplatform.googleapis.com/v1/publishers/google';
+
+// set `x-goog-api-key` header to API key for express mode
+function createExpressModeFetch(
+  apiKey: string,
+  customFetch?: FetchFunction,
+): FetchFunction {
+  return async (url, init) => {
+    const modifiedInit: RequestInit = {
+      ...init,
+      headers: {
+        ...(init?.headers ? normalizeHeaders(init.headers) : {}),
+        'x-goog-api-key': apiKey,
+      },
+    };
+    return (customFetch ?? fetch)(url.toString(), modifiedInit);
+  };
+}
+
+export interface GoogleVertexProvider extends ProviderV4 {
+  /**
+   * Creates a model for text generation.
+   */
+  (modelId: GoogleVertexModelId): LanguageModelV4;
+
+  languageModel: (modelId: GoogleVertexModelId) => LanguageModelV4;
+
+  /**
+   * Creates a model for image generation.
+   */
+  image(modelId: GoogleVertexImageModelId): ImageModelV4;
+
+  /**
+   * Creates a model for image generation.
+   */
+  imageModel(modelId: GoogleVertexImageModelId): ImageModelV4;
+
+  tools: typeof googleVertexTools;
+
+  /**
+   * @deprecated Use `embeddingModel` instead.
+   */
+  textEmbeddingModel(
+    modelId: GoogleVertexEmbeddingModelId,
+  ): GoogleVertexEmbeddingModel;
+
+  /**
+   * Creates a model for video generation.
+   */
+  video(modelId: GoogleVertexVideoModelId): Experimental_VideoModelV4;
+
+  /**
+   * Creates a model for video generation.
+   */
+  videoModel(modelId: GoogleVertexVideoModelId): Experimental_VideoModelV4;
+
+  /**
+   * Creates a model for speech generation (text-to-speech).
+   */
+  speech(modelId: GoogleVertexSpeechModelId): SpeechModelV4;
+
+  /**
+   * Creates a model for speech generation (text-to-speech).
+   */
+  speechModel(modelId: GoogleVertexSpeechModelId): SpeechModelV4;
+
+  /**
+   * Creates a model for transcription (speech-to-text).
+   */
+  transcription(
+    modelId: GoogleVertexTranscriptionModelId,
+  ): TranscriptionModelV4;
+
+  /**
+   * Creates a model for transcription (speech-to-text).
+   */
+  transcriptionModel(
+    modelId: GoogleVertexTranscriptionModelId,
+  ): TranscriptionModelV4;
+}
+
+export interface GoogleVertexProviderSettings {
+  /**
+   * Optional. The API key for the Google Cloud project. If provided, the
+   * provider will use express mode with API key authentication. Defaults to
+   * the value of the `GOOGLE_VERTEX_API_KEY` environment variable.
+   */
+  apiKey?: string;
+
+  /**
+   * Your Google Vertex location. Defaults to the environment variable `GOOGLE_VERTEX_LOCATION`.
+   */
+  location?: string;
+
+  /**
+   * Your Google Vertex project. Defaults to the environment variable `GOOGLE_VERTEX_PROJECT`.
+   */
+  project?: string;
+
+  /**
+   * Headers to use for requests. Can be:
+   * - A headers object
+   * - A Promise that resolves to a headers object
+   * - A function that returns a headers object
+   * - A function that returns a Promise of a headers object
+   */
+  headers?: Resolvable<Record<string, string | undefined>>;
+
+  /**
+   * Custom fetch implementation. You can use it as a middleware to intercept requests,
+   * or to provide a custom fetch implementation for e.g. testing.
+   */
+  fetch?: FetchFunction;
+
+  // for testing
+  generateId?: () => string;
+
+  /**
+   * Base URL for the Google Vertex API calls.
+   */
+  baseURL?: string;
+}
+
+/**
+ * Create a Google Vertex AI provider instance.
+ */
+export function createGoogleVertex(
+  options: GoogleVertexProviderSettings = {},
+): GoogleVertexProvider {
+  const apiKey = loadOptionalSetting({
+    settingValue: options.apiKey,
+    environmentVariableName: 'GOOGLE_VERTEX_API_KEY',
+  });
+
+  const loadGoogleVertexProject = () =>
+    loadSetting({
+      settingValue: options.project,
+      settingName: 'project',
+      environmentVariableName: 'GOOGLE_VERTEX_PROJECT',
+      description: 'Google Vertex project',
+    });
+
+  const loadGoogleVertexLocation = () =>
+    loadSetting({
+      settingValue: options.location,
+      settingName: 'location',
+      environmentVariableName: 'GOOGLE_VERTEX_LOCATION',
+      description: 'Google Vertex location',
+    });
+
+  const loadBaseURL = () => {
+    if (apiKey) {
+      return withoutTrailingSlash(options.baseURL) ?? EXPRESS_MODE_BASE_URL;
+    }
+
+    const region = loadGoogleVertexLocation();
+    const project = loadGoogleVertexProject();
+
+    const getHost = () => {
+      if (region === 'global') {
+        return 'aiplatform.googleapis.com';
+      } else if (region === 'eu' || region === 'us') {
+        return `aiplatform.${region}.rep.googleapis.com`;
+      } else {
+        return `${region}-aiplatform.googleapis.com`;
+      }
+    };
+
+    return (
+      withoutTrailingSlash(options.baseURL) ??
+      `https://${getHost()}/v1beta1/projects/${project}/locations/${region}/publishers/google`
+    );
+  };
+
+  const createConfig = (name: string): GoogleVertexConfig => {
+    const getHeaders = async () => {
+      const originalHeaders = await resolve(options.headers ?? {});
+      return withUserAgentSuffix(
+        originalHeaders,
+        `ai-sdk/google-vertex/${VERSION}`,
+      );
+    };
+
+    return {
+      provider: `google.vertex.${name}`,
+      headers: getHeaders,
+      fetch: apiKey
+        ? createExpressModeFetch(apiKey, options.fetch)
+        : options.fetch,
+      baseURL: loadBaseURL(),
+    };
+  };
+
+  const createChatModel = (modelId: GoogleVertexModelId) => {
+    return new GoogleLanguageModel(modelId, {
+      ...createConfig('chat'),
+      generateId: options.generateId ?? generateId,
+      supportedUrls: () => ({
+        '*': [
+          // HTTP URLs:
+          /^https?:\/\/.*$/,
+          // Google Cloud Storage URLs:
+          /^gs:\/\/.*$/,
+        ],
+      }),
+    });
+  };
+
+  const createEmbeddingModel = (modelId: GoogleVertexEmbeddingModelId) =>
+    new GoogleVertexEmbeddingModel(modelId, createConfig('embedding'));
+
+  const createImageModel = (modelId: GoogleVertexImageModelId) =>
+    new GoogleVertexImageModel(modelId, {
+      ...createConfig('image'),
+      generateId: options.generateId ?? generateId,
+    });
+
+  const createVideoModel = (modelId: GoogleVertexVideoModelId) =>
+    new GoogleVertexVideoModel(modelId, {
+      ...createConfig('video'),
+      generateId: options.generateId ?? generateId,
+    });
+
+  const createSpeechModel = (modelId: GoogleVertexSpeechModelId) =>
+    new GoogleSpeechModel(modelId, createConfig('speech'));
+
+  // Cloud Speech-to-Text reuses the Vertex auth headers from createConfig, but
+  // targets the Speech-to-Text API.
+  const createTranscriptionModel = (
+    modelId: GoogleVertexTranscriptionModelId,
+  ) => {
+    if (apiKey) {
+      throw new Error(
+        'Google Vertex transcription models do not support Express Mode API keys. Use standard Google Cloud credentials instead.',
+      );
+    }
+
+    const config = createConfig('transcription');
+    return new GoogleVertexTranscriptionModel(modelId, {
+      provider: config.provider,
+      headers: config.headers,
+      fetch: config.fetch,
+      project: loadGoogleVertexProject(),
+      location: loadGoogleVertexLocation(),
+    });
+  };
+
+  const provider = function (modelId: GoogleVertexModelId) {
+    if (new.target) {
+      throw new Error(
+        'The Google Vertex AI model function cannot be called with the new keyword.',
+      );
+    }
+
+    return createChatModel(modelId);
+  };
+
+  provider.specificationVersion = 'v4' as const;
+  provider.languageModel = createChatModel;
+  provider.embeddingModel = createEmbeddingModel;
+  provider.textEmbeddingModel = createEmbeddingModel;
+  provider.image = createImageModel;
+  provider.imageModel = createImageModel;
+  provider.video = createVideoModel;
+  provider.videoModel = createVideoModel;
+  provider.speech = createSpeechModel;
+  provider.speechModel = createSpeechModel;
+  provider.transcription = createTranscriptionModel;
+  provider.transcriptionModel = createTranscriptionModel;
+  provider.tools = googleVertexTools;
+
+  return provider;
+}

package/src/google-vertex-provider.ts

@@ -1,136 +1,24 @@
-import { GoogleGenerativeAILanguageModel } from '@ai-sdk/google/internal';
+import { loadOptionalSetting, resolve } from '@ai-sdk/provider-utils';
+import type { GoogleAuthOptions } from 'google-auth-library';
+import { createAuthTokenGenerator } from './google-vertex-auth-google-auth-library';
 import {
-  Experimental_VideoModelV3,
-  ImageModelV3,
-  LanguageModelV3,
-  ProviderV3,
-} from '@ai-sdk/provider';
-import {
-  FetchFunction,
-  generateId,
-  loadOptionalSetting,
-  loadSetting,
-  normalizeHeaders,
-  resolve,
-  Resolvable,
-  withoutTrailingSlash,
-  withUserAgentSuffix,
-} from '@ai-sdk/provider-utils';
-import { VERSION } from './version';
-import { GoogleVertexConfig } from './google-vertex-config';
-import { GoogleVertexEmbeddingModel } from './google-vertex-embedding-model';
-import { GoogleVertexEmbeddingModelId } from './google-vertex-embedding-options';
-import { GoogleVertexImageModel } from './google-vertex-image-model';
-import { GoogleVertexImageModelId } from './google-vertex-image-settings';
-import { GoogleVertexModelId } from './google-vertex-options';
-import { googleVertexTools } from './google-vertex-tools';
-import { GoogleVertexVideoModel } from './google-vertex-video-model';
-import { GoogleVertexVideoModelId } from './google-vertex-video-settings';
-
-const EXPRESS_MODE_BASE_URL =
-  'https://aiplatform.googleapis.com/v1/publishers/google';
-
-// set `x-goog-api-key` header to API key for express mode
-function createExpressModeFetch(
-  apiKey: string,
-  customFetch?: FetchFunction,
-): FetchFunction {
-  return async (url, init) => {
-    const modifiedInit: RequestInit = {
-      ...init,
-      headers: {
-        ...(init?.headers ? normalizeHeaders(init.headers) : {}),
-        'x-goog-api-key': apiKey,
-      },
-    };
-    return (customFetch ?? fetch)(url.toString(), modifiedInit);
-  };
-}
-
-export interface GoogleVertexProvider extends ProviderV3 {
-  /**
-   * Creates a model for text generation.
-   */
-  (modelId: GoogleVertexModelId): LanguageModelV3;
-
-  languageModel: (modelId: GoogleVertexModelId) => LanguageModelV3;
-
-  /**
-   * Creates a model for image generation.
-   */
-  image(modelId: GoogleVertexImageModelId): ImageModelV3;
-
-  /**
-   * Creates a model for image generation.
-   */
-  imageModel(modelId: GoogleVertexImageModelId): ImageModelV3;
-
-  tools: typeof googleVertexTools;
-
-  /**
-   * @deprecated Use `embeddingModel` instead.
-   */
-  textEmbeddingModel(
-    modelId: GoogleVertexEmbeddingModelId,
-  ): GoogleVertexEmbeddingModel;
-
-  /**
-   * Creates a model for video generation.
-   */
-  video(modelId: GoogleVertexVideoModelId): Experimental_VideoModelV3;
-
-  /**
-   * Creates a model for video generation.
-   */
-  videoModel(modelId: GoogleVertexVideoModelId): Experimental_VideoModelV3;
+  createGoogleVertex as createGoogleVertexOriginal,
+  type GoogleVertexProvider,
+  type GoogleVertexProviderSettings as GoogleVertexProviderSettingsOriginal,
+} from './google-vertex-provider-base';
+export interface GoogleVertexProviderSettings extends GoogleVertexProviderSettingsOriginal {
+  /**
+   * Optional. The Authentication options provided by google-auth-library.
+   * Complete list of authentication options is documented in the
+   * GoogleAuthOptions interface:
+   * https://github.com/googleapis/google-auth-library-nodejs/blob/main/src/auth/googleauth.ts.
+   */
+  googleAuthOptions?: GoogleAuthOptions;
 }
 
-export interface GoogleVertexProviderSettings {
-  /**
-   * Optional. The API key for the Google Cloud project. If provided, the
-   * provider will use express mode with API key authentication. Defaults to
-   * the value of the `GOOGLE_VERTEX_API_KEY` environment variable.
-   */
-  apiKey?: string;
-
-  /**
-   * Your Google Vertex location. Defaults to the environment variable `GOOGLE_VERTEX_LOCATION`.
-   */
-  location?: string;
-
-  /**
-   * Your Google Vertex project. Defaults to the environment variable `GOOGLE_VERTEX_PROJECT`.
-   */
-  project?: string;
-
-  /**
-   * Headers to use for requests. Can be:
-   * - A headers object
-   * - A Promise that resolves to a headers object
-   * - A function that returns a headers object
-   * - A function that returns a Promise of a headers object
-   */
-  headers?: Resolvable<Record<string, string | undefined>>;
-
-  /**
-   * Custom fetch implementation. You can use it as a middleware to intercept requests,
-   * or to provide a custom fetch implementation for e.g. testing.
-   */
-  fetch?: FetchFunction;
-
-  // for testing
-  generateId?: () => string;
-
-  /**
-   * Base URL for the Google Vertex API calls.
-   */
-  baseURL?: string;
-}
+export type { GoogleVertexProvider };
 
-/**
- * Create a Google Vertex AI provider instance.
- */
-export function createVertex(
+export function createGoogleVertex(
   options: GoogleVertexProviderSettings = {},
 ): GoogleVertexProvider {
   const apiKey = loadOptionalSetting({
@@ -138,108 +26,30 @@ export function createVertex(
     environmentVariableName: 'GOOGLE_VERTEX_API_KEY',
   });
 
-  const loadVertexProject = () =>
-    loadSetting({
-      settingValue: options.project,
-      settingName: 'project',
-      environmentVariableName: 'GOOGLE_VERTEX_PROJECT',
-      description: 'Google Vertex project',
-    });
-
-  const loadVertexLocation = () =>
-    loadSetting({
-      settingValue: options.location,
-      settingName: 'location',
-      environmentVariableName: 'GOOGLE_VERTEX_LOCATION',
-      description: 'Google Vertex location',
-    });
-
-  const loadBaseURL = () => {
-    if (apiKey) {
-      return withoutTrailingSlash(options.baseURL) ?? EXPRESS_MODE_BASE_URL;
-    }
-
-    const region = loadVertexLocation();
-    const project = loadVertexProject();
-
-    // For global region, use aiplatform.googleapis.com directly
-    // For other regions, use region-aiplatform.googleapis.com
-    const baseHost = `${region === 'global' ? '' : region + '-'}aiplatform.googleapis.com`;
-
-    return (
-      withoutTrailingSlash(options.baseURL) ??
-      `https://${baseHost}/v1beta1/projects/${project}/locations/${region}/publishers/google`
-    );
-  };
-
-  const createConfig = (name: string): GoogleVertexConfig => {
-    const getHeaders = async () => {
-      const originalHeaders = await resolve(options.headers ?? {});
-      return withUserAgentSuffix(
-        originalHeaders,
-        `ai-sdk/google-vertex/${VERSION}`,
-      );
-    };
-
-    return {
-      provider: `google.vertex.${name}`,
-      headers: getHeaders,
-      fetch: apiKey
-        ? createExpressModeFetch(apiKey, options.fetch)
-        : options.fetch,
-      baseURL: loadBaseURL(),
-    };
-  };
-
-  const createChatModel = (modelId: GoogleVertexModelId) => {
-    return new GoogleGenerativeAILanguageModel(modelId, {
-      ...createConfig('chat'),
-      generateId: options.generateId ?? generateId,
-      supportedUrls: () => ({
-        '*': [
-          // HTTP URLs:
-          /^https?:\/\/.*$/,
-          // Google Cloud Storage URLs:
-          /^gs:\/\/.*$/,
-        ],
-      }),
-    });
-  };
-
-  const createEmbeddingModel = (modelId: GoogleVertexEmbeddingModelId) =>
-    new GoogleVertexEmbeddingModel(modelId, createConfig('embedding'));
-
-  const createImageModel = (modelId: GoogleVertexImageModelId) =>
-    new GoogleVertexImageModel(modelId, {
-      ...createConfig('image'),
-      generateId: options.generateId ?? generateId,
-    });
-
-  const createVideoModel = (modelId: GoogleVertexVideoModelId) =>
-    new GoogleVertexVideoModel(modelId, {
-      ...createConfig('video'),
-      generateId: options.generateId ?? generateId,
-    });
-
-  const provider = function (modelId: GoogleVertexModelId) {
-    if (new.target) {
-      throw new Error(
-        'The Google Vertex AI model function cannot be called with the new keyword.',
-      );
-    }
-
-    return createChatModel(modelId);
-  };
-
-  provider.specificationVersion = 'v3' as const;
-  provider.languageModel = createChatModel;
-  provider.embeddingModel = createEmbeddingModel;
-  provider.textEmbeddingModel = createEmbeddingModel;
-  provider.image = createImageModel;
-  provider.imageModel = createImageModel;
-  provider.video = createVideoModel;
-  provider.videoModel = createVideoModel;
-  provider.tools = googleVertexTools;
-
-  return provider;
+  if (apiKey) {
+    return createGoogleVertexOriginal(options);
+  }
+
+  const googleAuthOptions =
+    options.project == null
+      ? options.googleAuthOptions
+      : {
+          projectId: options.project,
+          ...options.googleAuthOptions,
+        };
+
+  const generateAuthToken = createAuthTokenGenerator(googleAuthOptions);
+
+  return createGoogleVertexOriginal({
+    ...options,
+    headers: async () => ({
+      Authorization: `Bearer ${await generateAuthToken()}`,
+      ...(await resolve(options.headers)),
+    }),
+  });
 }
+
+/**
+ * Default Google Vertex AI provider instance.
+ */
+export const googleVertex = createGoogleVertex();

package/src/google-vertex-speech-model-options.ts

@@ -0,0 +1,11 @@
+import type { GoogleSpeechModelOptions } from '@ai-sdk/google';
+
+// https://docs.cloud.google.com/text-to-speech/docs/gemini-tts
+export type GoogleVertexSpeechModelId =
+  | 'gemini-2.5-flash-tts'
+  | 'gemini-2.5-pro-tts'
+  | 'gemini-2.5-flash-lite-preview-tts'
+  | 'gemini-3.1-flash-tts-preview'
+  | (string & {});
+
+export type GoogleVertexSpeechModelOptions = GoogleSpeechModelOptions;

package/src/google-vertex-transcription-model-options.ts

@@ -0,0 +1,46 @@
+import { z } from 'zod/v4';
+
+// Google Cloud Speech-to-Text transcription models (Speech-to-Text v2).
+// https://docs.cloud.google.com/speech-to-text/docs/transcription-model
+export type GoogleVertexTranscriptionModelId =
+  | 'chirp_2'
+  | 'chirp_3'
+  | 'telephony'
+  | (string & {});
+
+export const googleVertexTranscriptionProviderOptionsSchema = z.object({
+  /**
+   * BCP-47 language codes to recognize (e.g. `['en-US']`), or `['auto']` to let
+   * Chirp auto-detect the spoken language. Defaults to `['auto']`. For
+   * `telephony`, pass a supported explicit language code.
+   */
+  languageCodes: z.array(z.string()).optional(),
+
+  /**
+   * Whether to add punctuation to the transcript. Defaults to `true`.
+   */
+  enableAutomaticPunctuation: z.boolean().optional(),
+
+  /**
+   * Whether to include word-level timestamps. Defaults to `true` so the
+   * transcription result can include segments.
+   *
+   * Enabling word-level timestamps can reduce transcription quality and speed
+   * for Chirp models.
+   */
+  enableWordTimeOffsets: z.boolean().optional(),
+
+  /**
+   * The Cloud Speech-to-Text region for the request (e.g. `'us'`, `'eu'`,
+   * `'us-central1'`). Defaults to the provider `location`.
+   *
+   * Note: Speech-to-Text regions differ from Vertex AI regions. Chirp is only
+   * available in specific Speech-to-Text regions and is not available in the
+   * `global` location.
+   */
+  region: z.string().optional(),
+});
+
+export type GoogleVertexTranscriptionModelOptions = z.infer<
+  typeof googleVertexTranscriptionProviderOptionsSchema
+>;