Gemini 3.8 Live Extended Thinking:支持扩展推理的音频到音频模型
英文标题:🔶 Gemini 3.8 Live Extended Thinking audio-to-audio
Gemini 3.8 Live Extended Thinking 是一款面向实时语音交互的高推理能力音频到音频模型,适用于复杂、多步骤问题求解。该模型可在持续流式输出音频回复的同时,执行后台推理和异步工具调用。开发者需要调整客户端状态管理:`turnComplete: true` 不再代表模型已空闲,而应通过 `interaction_status` 判断处理状态。该模型仅支持非阻塞式异步函数调用,并可通过 `thinking_config` 设置低、中、高三档后台推理级别。
256.5 KB
中文
Gemini 3.8 Live Extended Thinking
Gemini 3.8 Live Extended Thinking 是一款高推理能力的音频到音频模型,建议在实时语音交互中需要更强后台推理能力来解决复杂、多步骤问题时使用。它能够在持续流式输出音频回复的同时,处理后台推理和异步工具调用。
文档
如需完整了解功能与能力,请访问 Live API 指南。
gemini-3.8-live-extended-thinking
升级至 Gemini 3.8 Live Extended Thinking
Gemini 3.8 Live Extended Thinking 在实时音频会话中引入了后台推理。集成该模型时,请更新客户端状态管理,以处理异步推理信号:
- 异步推理协议:与使用异步推理的模型交互时,
turnComplete: true不再表示模型处于空闲状态。服务器可能仍在处理后台推理或工具调用。在收到turnComplete: true后,客户端必须继续监听后续服务器消息,例如工具调用或音频帧。 - 监控
interaction_status:使用传入服务器消息中的interaction_status字段判断服务器当前状态:IN_PROGRESS:服务器正在积极处理用户输入、执行后台推理,或等待异步工具调用的响应。后续可能继续产生模型输出或工具调用。IDLE:服务器已完成全部处理、推理和工具调用。会话处于空闲状态,正在等待用户输入。
- 异步函数调用:仅支持异步非阻塞执行(
behavior: NON_BLOCKING)。不支持同步阻塞模式,使用该模式会返回硬错误。不支持函数调度配置。 - 思考配置:通过设置配置中的
thinking_config配置后台推理(thinking_level可选值为low、medium或high)。请注意,不支持MINIMAL。 - 客户端内容更新:在整个会话生命周期内均支持
send_client_content,且必须显式指定角色(user或model)。设置turn_complete=true会立即中断正在进行的生成。 - 主动音频:永久启用。设置
proactive_audio: false会返回错误。
如需比较所有 Live API 模型的功能,请参阅模型对比表。
最后更新:2026-09-15 UTC。
English / 原文
Send feedback
Gemini 3.8 Live Extended Thinking
Gemini 3.8 Live Extended Thinking is our high-reasoning audio-to-audio model recommended when higher background reasoning is required for complex, multi-step problem solving during real-time voice interactions. It processes background reasoning and asynchronous tool calls while streaming continuous audio responses.
Documentation
Visit the Live API guide for full coverage of features and capabilities.
gemini-3.8-live-extended-thinking
Upgrading to Gemini 3.8 Live Extended Thinking
Gemini 3.8 Live Extended Thinking introduces background reasoning during live audio sessions. When integrating this model, update your client state management to handle asynchronous reasoning signals:
- Asynchronous reasoning protocol : When interacting with models that use asynchronous reasoning,
turnComplete: trueno longer indicates that the model is idle. The server may continue processing background reasoning or tool calls. Your client must continue listening for subsequent server messages (such as tool calls or audio frames) afterturnComplete: truearrives. - Monitoring
interaction_status: Use theinteraction_statusfield on incoming server messages to determine current server state:IN_PROGRESS: The server is actively processing user input, running background reasoning, or awaiting responses for asynchronous tool calls. Additional model output or tool calls may follow.IDLE: The server has finished all processing, reasoning, and tool calls. The session is idle and waiting for user input.
- Asynchronous function calling : Only asynchronous non-blocking execution (
behavior: NON_BLOCKING) is supported. Synchronous blocking mode is not supported and returns a hard error. Function scheduling configurations are not supported. - Thinking configuration : Configure background reasoning using
thinking_configin your setup configuration (thinking_level:low,medium, orhigh). Note thatMINIMALis not supported. - Client content updates :
send_client_contentis supported throughout the entire session lifecycle with explicit roles (userormodel). Settingturn_complete=trueimmediately interrupts active generation. - Proactive audio : Permanently enabled. Setting
proactive_audio: falsereturns an error.
For a feature comparison across all Live API models, see the Model comparison table.
Send feedback
Except as otherwise noted, the content of this page is licensed under the Creative Commons Attribution 4.0 License, and code samples are licensed under the Apache 2.0 License. For details, see the Google Developers Site Policies. Java is a registered trademark of Oracle and/or its affiliates.
Last updated 2026-09-15 UTC.
Need to tell us more? [[["Easy to understand","easyToUnderstand","thumb-up"],["Solved my problem","solvedMyProblem","thumb-up"],["Other","otherUp","thumb-up"]],[["Missing the information I need","missingTheInformationINeed","thumb-down"],["Too complicated / too many steps","tooComplicatedTooManySteps","thumb-down"],["Out of date","outOfDate","thumb-down"],["Samples / code issue","samplesCodeIssue","thumb-down"],["Other","otherDown","thumb-down"]],["Last updated 2026-09-15 UTC."],[],[]]