chore(js/plugins/google-genai): Migrate to V2 plugin.

js/plugins/google-genai/src/googleai/embedder.ts

@@ -20,10 +20,10 @@ import {
   embedderActionMetadata,
   EmbedderInfo,
   EmbedderReference,
-  Genkit,
   z,
 } from 'genkit';
 import { embedderRef } from 'genkit/embedder';
+import { embedder as pluginEmbedder } from 'genkit/plugin';
 import { embedContent } from './client.js';
 import {
   EmbedContentRequest,
@@ -122,43 +122,40 @@ export function listActions(models: Model[]): ActionMetadata[] {
   );
 }
 
-export function defineKnownModels(ai: Genkit, options?: GoogleAIPluginOptions) {
-  for (const name of Object.keys(KNOWN_MODELS)) {
-    defineEmbedder(ai, name, options);
-  }
+export function listKnownModels(options?: GoogleAIPluginOptions) {
+  return Object.keys(KNOWN_MODELS).map((name) => defineEmbedder(name, options));
 }
 
 export function defineEmbedder(
-  ai: Genkit,
   name: string,
   pluginOptions?: GoogleAIPluginOptions
 ): EmbedderAction {
   checkApiKey(pluginOptions?.apiKey);
   const ref = model(name);
 
-  return ai.defineEmbedder(
+  return pluginEmbedder(
     {
       name: ref.name,
       configSchema: ref.configSchema,
       info: ref.info,
     },
-    async (input, reqOptions) => {
+    async (request, _) => {
       const embedApiKey = calculateApiKey(
         pluginOptions?.apiKey,
-        reqOptions?.apiKey
+        request.options?.apiKey
       );
-      const embedVersion = reqOptions?.version || extractVersion(ref);
+      const embedVersion = request.options?.version || extractVersion(ref);
 
       const embeddings = await Promise.all(
-        input.map(async (doc) => {
+        request.input.map(async (doc) => {
           const response = await embedContent(embedApiKey, embedVersion, {
-            taskType: reqOptions?.taskType,
-            title: reqOptions?.title,
+            taskType: request.options?.taskType,
+            title: request.options?.title,
             content: {
               role: '',
               parts: [{ text: doc.text }],
             },
-            outputDimensionality: reqOptions?.outputDimensionality,
+            outputDimensionality: request.options?.outputDimensionality,
           } as EmbedContentRequest);
           const values = response.embedding.values;
           return { embedding: values };

js/plugins/google-genai/src/googleai/gemini.ts

@@ -14,13 +14,7 @@
  * limitations under the License.
  */
 
-import {
-  ActionMetadata,
-  Genkit,
-  GenkitError,
-  modelActionMetadata,
-  z,
-} from 'genkit';
+import { ActionMetadata, GenkitError, modelActionMetadata, z } from 'genkit';
 import {
   GenerationCommonConfigDescriptions,
   GenerationCommonConfigSchema,
@@ -32,6 +26,7 @@ import {
   modelRef,
 } from 'genkit/model';
 import { downloadRequestMedia } from 'genkit/model/middleware';
+import { model as pluginModel } from 'genkit/plugin';
 import { runInNewSpan } from 'genkit/tracing';
 import {
   fromGeminiCandidate,
@@ -434,17 +429,16 @@ export function listActions(models: Model[]): ActionMetadata[] {
   );
 }
 
-export function defineKnownModels(ai: Genkit, options?: GoogleAIPluginOptions) {
-  for (const name of Object.keys(KNOWN_MODELS)) {
-    defineModel(ai, name, options);
-  }
+export function listKnownModels(options?: GoogleAIPluginOptions) {
+  return Object.keys(KNOWN_MODELS).map((name: string) =>
+    defineModel(name, options)
+  );
 }
 
 /**
  * Defines a new GoogleAI Gemini model.
  */
 export function defineModel(
-  ai: Genkit,
   name: string,
   pluginOptions?: GoogleAIPluginOptions
 ): ModelAction {
@@ -482,9 +476,8 @@ export function defineModel(
     );
   }
 
-  return ai.defineModel(
+  return pluginModel(
     {
-      apiVersion: 'v2',
       name: ref.name,
       ...ref.info,
       configSchema: ref.configSchema,
@@ -660,7 +653,6 @@ export function defineModel(
       // API params as for input.
       return pluginOptions?.experimental_debugTraces
         ? await runInNewSpan(
-            ai.registry,
             {
               metadata: {
                 name: streamingRequested ? 'sendMessageStream' : 'sendMessage',

js/plugins/google-genai/src/googleai/imagen.ts

@@ -20,7 +20,6 @@ import {
   MessageData,
   modelActionMetadata,
   z,
-  type Genkit,
 } from 'genkit';
 import {
   getBasicUsageStats,
@@ -30,6 +29,7 @@ import {
   type ModelInfo,
   type ModelReference,
 } from 'genkit/model';
+import { model as pluginModel } from 'genkit/plugin';
 import { imagenPredict } from './client.js';
 import type {
   ClientOptions,
@@ -169,14 +169,13 @@ export function listActions(models: Model[]): ActionMetadata[] {
     });
 }
 
-export function defineKnownModels(ai: Genkit, options?: GoogleAIPluginOptions) {
-  for (const name of Object.keys(KNOWN_MODELS)) {
-    defineModel(ai, name, options);
-  }
+export function listKnownModels(options?: GoogleAIPluginOptions) {
+  return Object.keys(KNOWN_MODELS).map((name: string) =>
+    defineModel(name, options)
+  );
 }
 
 export function defineModel(
-  ai: Genkit,
   name: string,
   pluginOptions?: GoogleAIPluginOptions
 ): ModelAction {
@@ -187,9 +186,8 @@ export function defineModel(
     baseUrl: pluginOptions?.baseUrl,
   };
 
-  return ai.defineModel(
+  return pluginModel(
     {
-      apiVersion: 'v2',
       name: ref.name,
       ...ref.info,
       configSchema: ref.configSchema,

js/plugins/google-genai/src/googleai/index.ts

@@ -14,15 +14,13 @@
  * limitations under the License.
  */
 
-import {
-  ActionMetadata,
-  EmbedderReference,
-  Genkit,
-  ModelReference,
-  z,
-} from 'genkit';
+import { ActionMetadata, EmbedderReference, ModelReference, z } from 'genkit';
 import { logger } from 'genkit/logging';
-import { GenkitPlugin, genkitPlugin } from 'genkit/plugin';
+import {
+  GenkitPluginV2,
+  ResolvableAction,
+  genkitPluginV2,
+} from 'genkit/plugin';
 import { ActionType } from 'genkit/registry';
 import { extractErrMsg } from '../common/utils.js';
 import { listModels } from './client.js';
@@ -42,41 +40,41 @@ export { type GeminiConfig, type GeminiTtsConfig } from './gemini.js';
 export { type ImagenConfig } from './imagen.js';
 export { type GoogleAIPluginOptions };
 
-async function initializer(ai: Genkit, options?: GoogleAIPluginOptions) {
-  imagen.defineKnownModels(ai, options);
-  gemini.defineKnownModels(ai, options);
-  embedder.defineKnownModels(ai, options);
-  veo.defineKnownModels(ai, options);
+async function initializer(options?: GoogleAIPluginOptions) {
+  return [
+    ...imagen.listKnownModels(options),
+    ...gemini.listKnownModels(options),
+    ...embedder.listKnownModels(options),
+    ...veo.listKnownModels(options),
+  ];
 }
 
 async function resolver(
-  ai: Genkit,
   actionType: ActionType,
   actionName: string,
   options: GoogleAIPluginOptions
-) {
+): Promise<ResolvableAction | undefined> {
   switch (actionType) {
     case 'model':
       if (veo.isVeoModelName(actionName)) {
-        // no-op (not gemini)
+        return undefined;
       } else if (imagen.isImagenModelName(actionName)) {
-        imagen.defineModel(ai, actionName, options);
+        return await imagen.defineModel(actionName, options);
       } else {
         // gemini, tts, gemma, unknown models
-        gemini.defineModel(ai, actionName, options);
+        return await gemini.defineModel(actionName, options);
       }
       break;
     case 'background-model':
       if (veo.isVeoModelName(actionName)) {
-        veo.defineModel(ai, actionName, options);
+        return await veo.defineModel(actionName, options);
       }
       break;
     case 'embedder':
-      embedder.defineEmbedder(ai, actionName, options);
+      return await embedder.defineEmbedder(actionName, options);
       break;
-    default:
-    // no-op
   }
+  return undefined;
 }
 
 async function listActions(
@@ -104,23 +102,25 @@ async function listActions(
 /**
  * Google Gemini Developer API plugin.
  */
-export function googleAIPlugin(options?: GoogleAIPluginOptions): GenkitPlugin {
+export function googleAIPlugin(
+  options?: GoogleAIPluginOptions
+): GenkitPluginV2 {
   let listActionsCache;
-  return genkitPlugin(
-    'googleai',
-    async (ai: Genkit) => await initializer(ai, options),
-    async (ai: Genkit, actionType: ActionType, actionName: string) =>
-      await resolver(ai, actionType, actionName, options || {}),
-    async () => {
+  return genkitPluginV2({
+    name: 'googleai',
+    init: async () => await initializer(options),
+    resolve: async (actionType: ActionType, actionName: string) =>
+      await resolver(actionType, actionName, options || {}),
+    list: async () => {
       if (listActionsCache) return listActionsCache;
       listActionsCache = await listActions(options);
       return listActionsCache;
-    }
-  );
+    },
+  });
 }
 
 export type GoogleAIPlugin = {
-  (pluginOptions?: GoogleAIPluginOptions): GenkitPlugin;
+  (pluginOptions?: GoogleAIPluginOptions): GenkitPluginV2;
   model(
     name: gemini.KnownGemmaModels | (gemini.GemmaModelName & {}),
     config: gemini.GemmaConfig

js/plugins/google-genai/src/googleai/veo.ts

@@ -20,7 +20,6 @@ import {
   Operation,
   modelActionMetadata,
   z,
-  type Genkit,
 } from 'genkit';
 import {
   BackgroundModelAction,
@@ -29,6 +28,7 @@ import {
   type ModelInfo,
   type ModelReference,
 } from 'genkit/model';
+import { backgroundModel as pluginBackgroundModel } from 'genkit/plugin';
 import { veoCheckOperation, veoPredict } from './client.js';
 import {
   ClientOptions,
@@ -156,17 +156,16 @@ export function listActions(models: Model[]): ActionMetadata[] {
   );
 }
 
-export function defineKnownModels(ai: Genkit, options?: GoogleAIPluginOptions) {
-  for (const name of Object.keys(KNOWN_MODELS)) {
-    defineModel(ai, name, options);
-  }
+export function listKnownModels(options?: GoogleAIPluginOptions) {
+  return Object.keys(KNOWN_MODELS).map((name: string) =>
+    defineModel(name, options)
+  );
 }
 
 /**
  * Defines a new GoogleAI Veo model.
  */
 export function defineModel(
-  ai: Genkit,
   name: string,
   pluginOptions?: GoogleAIPluginOptions
 ): BackgroundModelAction<VeoConfigSchemaType> {
@@ -176,7 +175,7 @@ export function defineModel(
     baseUrl: pluginOptions?.baseUrl,
   };
 
-  return ai.defineBackgroundModel({
+  return pluginBackgroundModel({
     name: ref.name,
     ...ref.info,
     configSchema: ref.configSchema,