`@anvia/gemini`

Gemini’s provider adapter connects Anvia to the Gemini Developer API or Vertex AI. It includes completion, embeddings, native Gemini image generation, Imagen generation, transcription, and model listing.


Support	First-party
Version	`1.1.1`
Runtime	ESM, server-side JavaScript
Peer	Matching `@anvia/core` stable release

Install

bash

pnpm add @anvia/gemini @anvia/core

Create a Gemini agent

import { Agent } from '@anvia/core'
import { GeminiClient } from '@anvia/gemini'

const gemini = new GeminiClient({
  apiKey: process.env.GEMINI_API_KEY!,
})

const agent = new Agent({
  id: 'assistant',
  model: gemini.completionModel({
      modelId: 'gemini-3.6-flash'
  }),
})

const result = await agent.generate({
    prompt: 'Describe this system in three bullets.'
})

if (result.type === 'response') {
  console.log(result.output)
}

Capabilities

Capability	Factory	Default
Streaming completion	`completionModel({ modelId })`	Explicit model
Dense embeddings	`embeddingModel({ modelId })`	Explicit model
Gemini-native images	`imageGenerationModel({ api: 'generateContent', modelId })`	Explicit model
Imagen images	`imageGenerationModel({ api: 'generateImages', modelId })`	Explicit model
Audio transcription	`transcriptionModel({ modelId })`	Explicit model
Model inventory	`listModels()`	Provider model list

Gemini-native images run through generateContent; Imagen uses generateImages. One discriminated factory supports both provider APIs and model families. Audio generation is not implemented.

Common patterns

Use Vertex AI

const vertex = new GeminiClient({
  vertexAi: {
    projectId: 'my-gcp-project',
    location: 'us-central1',
  },
})

const model = vertex.completionModel({
    modelId: 'gemini-3.6-flash'
})

Vertex mode uses Google Application Default Credentials unless googleAuthOptions supplies another trusted configuration.

Tune embeddings for retrieval

const queryEmbeddings = gemini.embeddingModel({
    modelId: 'gemini-embedding-001',
    taskType: 'RETRIEVAL_QUERY',
    dimensions: 768,
    maxBatchSize: 32
})

Use matching dimensions and a compatible task configuration when indexing and querying the same vector collection. title is intended for document-oriented embedding tasks.

Compatibility

@anvia/gemini is ESM and uses @google/genai. The client options are a discriminated union: API-key mode cannot include vertexAi, and Vertex mode cannot include apiKey. A preconfigured GoogleGenAI client is the third mutually exclusive form.

@anvia/gemini ​

Install ​

Create a Gemini agent ​

Capabilities ​

Common patterns ​

Use Vertex AI ​

Tune embeddings for retrieval ​

Compatibility ​

Continue ​