OpenAI - AgentScope Java

agentscope-extensions-model-openai integrates OpenAI Chat Completions-style models. It is also the module to use for OpenAI-compatible endpoints such as DeepSeek, GLM, Kimi, MiniMax, and similar services when their wire format follows the OpenAI API.

Add the dependency

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

ModelRegistry

Set OPENAI_API_KEY, then use the openai:<model> id:

ReActAgent agent = ReActAgent.builder()
    .name("assistant")
    .model("openai:gpt-4.1-mini") // Resolved internally by ModelRegistry.resolve(modelId)
    .build();

Explicit builder

Use the builder when you need a custom endpoint, formatter, transport, or generation options:

import io.agentscope.extensions.model.openai.OpenAIChatModel;

OpenAIChatModel model = OpenAIChatModel.builder()
    .apiKey(System.getenv("OPENAI_API_KEY"))
    .modelName("gpt-4.1-mini")
    .stream(true)
    .build();

Spring Boot

Spring Boot applications can use the OpenAI starter:

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

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