Parameterize default in-cloud model name

Author: erikeldridge

common/api-review/vertexai.api.md

@@ -326,6 +326,7 @@ export interface GenerativeContentBlob {
 export class GenerativeModel extends VertexAIModel {
     constructor(vertexAI: VertexAI, modelParams: ModelParams, requestOptions?: RequestOptions);
     countTokens(request: CountTokensRequest | string | Array<string | Part>): Promise<CountTokensResponse>;
+    static DEFAULT_HYBRID_IN_CLOUD_MODEL: string;
     generateContent(request: GenerateContentRequest | string | Array<string | Part>): Promise<GenerateContentResult>;
     generateContentStream(request: GenerateContentRequest | string | Array<string | Part>): Promise<GenerateContentStreamResult>;
     // (undocumented)
@@ -418,8 +419,8 @@ export enum HarmSeverity {
 
 // @public
 export interface HybridParams {
-    mode?: InferenceMode;
-    onCloudParams?: ModelParams;
+    inCloudParams?: ModelParams;
+    mode: InferenceMode;
     onDeviceParams?: AILanguageModelCreateOptionsWithSystemPrompt;
 }
 
@@ -509,7 +510,7 @@ export interface ImagenSafetySettings {
 
 // @public
 export enum InferenceMode {
-    ONLY_ON_CLOUD = "ONLY_ON_CLOUD",
+    ONLY_IN_CLOUD = "ONLY_IN_CLOUD",
     ONLY_ON_DEVICE = "ONLY_ON_DEVICE",
     PREFER_ON_DEVICE = "PREFER_ON_DEVICE"
 }

packages/vertexai/src/api.test.ts

@@ -106,7 +106,15 @@ describe('Top level API', () => {
     expect(genModel).to.be.an.instanceOf(GenerativeModel);
     expect(genModel.model).to.equal('publishers/google/models/my-model');
   });
-  it('getGenerativeModel with HybridParams sets the model', () => {
+  it('getGenerativeModel with HybridParams sets a default model', () => {
+    const genModel = getGenerativeModel(fakeVertexAI, {
+      mode: InferenceMode.ONLY_ON_DEVICE
+    });
+    expect(genModel.model).to.equal(
+      `publishers/google/models/${GenerativeModel.DEFAULT_HYBRID_IN_CLOUD_MODEL}`
+    );
+  });
+  it('getGenerativeModel with HybridParams honors a model override', () => {
     const genModel = getGenerativeModel(fakeVertexAI, {
       mode: InferenceMode.ONLY_IN_CLOUD,
       inCloudParams: { model: 'my-model' }

packages/vertexai/src/api.ts

@@ -79,7 +79,7 @@ export function getGenerativeModel(
   let inCloudParams: ModelParams;
   if (hybridParams.mode) {
     inCloudParams = hybridParams.inCloudParams || {
-      model: 'gemini-2.0-flash-lite'
+      model: GenerativeModel.DEFAULT_HYBRID_IN_CLOUD_MODEL
     };
   } else {
     inCloudParams = modelParams as ModelParams;

packages/vertexai/src/models/generative-model.ts

@@ -49,6 +49,10 @@ import { VertexAIModel } from './vertexai-model';
  * @public
  */
 export class GenerativeModel extends VertexAIModel {
+  /**
+   * Defines the name of the default in-cloud model to use for hybrid inference.
+   */
+  static DEFAULT_HYBRID_IN_CLOUD_MODEL = 'gemini-2.0-flash-lite';
   generationConfig: GenerationConfig;
   safetySettings: SafetySetting[];
   requestOptions?: RequestOptions;

packages/vertexai/src/types/enums.ts

@@ -243,7 +243,6 @@ export enum Modality {
 
 /**
  * Determines whether inference happens on-device or in-cloud.
- * @public
  */
 export enum InferenceMode {
   /**

packages/vertexai/src/types/requests.ts

@@ -216,14 +216,13 @@ export interface FunctionCallingConfig {
 }
 
 /**
- * Configures on-device and in-cloud inference.
- * @public
+ * Toggles hybrid inference.
  */
 export interface HybridParams {
   /**
-   * Optional. Specifies on-device or in-cloud inference. Defaults to prefer on-device.
+   * Specifies on-device or in-cloud inference. Defaults to prefer on-device.
    */
-  mode?: InferenceMode;
+  mode: InferenceMode;
   /**
    * Optional. Specifies advanced params for on-device inference.
    */