Gemini - AgentScope Java

agentscope-extensions-model-gemini integrates Google Gemini models through the Gemini API and supports the Vertex AI path through explicit configuration.

Add the dependency

<dependency>
    <groupId>io.agentscope</groupId>
    <artifactId>agentscope-extensions-model-gemini</artifactId>
    <version>${agentscope.version}</version>
</dependency>

ModelRegistry

Set GEMINI_API_KEY, then use the gemini:<model> id:

ReActAgent agent = ReActAgent.builder()
    .name("assistant")
    .model("gemini:gemini-2.0-flash") // Resolved internally by ModelRegistry.resolve(modelId)
    .build();

Explicit builder

Use the builder when you need custom API settings, Vertex AI credentials, formatter, transport, or generation options:

import io.agentscope.extensions.model.gemini.GeminiChatModel;

GeminiChatModel model = GeminiChatModel.builder()
    .apiKey(System.getenv("GEMINI_API_KEY"))
    .modelName("gemini-2.0-flash")
    .streamEnabled(true)
    .build();

Spring Boot

Spring Boot applications can use the Gemini starter:

<dependency>
    <groupId>io.agentscope</groupId>
    <artifactId>agentscope-gemini-spring-boot-starter</artifactId>
    <version>${agentscope.version}</version>
</dependency>

Full builder options, formatters, credentials, and registry context details are covered in Model.