Gemini 3.8 Live

默认的低延迟语音智能体模型，用于无推理延迟的实时对话。

Googlegemini-3.8-live2026-09-15报告数据错误

规格

上下文: —
最大输出: 65.5K
输入价: —
输出价: —
发布日期: 2026-09-15

上下文

上下文: —
最大输入: 131.1K
最大输出: 65.5K

价格

官方未公布

输入输出

输入: 文本图像音频视频
输出: 文本音频

API

接口类型: realtime

思考

支持思考: 支持
思考等级: —

源信息

状态: 可用
发布日期: 2026-09-15
知识截止: —

能力

已确认: tools

调用方式

googlemodel = gemini-3.8-live

JSON

gemini-3.8-live.json

{
  "id": "gemini-3.8-live",
  "object": "model",
  "created": 1789430400,
  "owned_by": "google",
  "name": "Gemini 3.8 Live",
  "api": {
    "types": [
      "realtime"
    ]
  },
  "limits": {
    "context": null,
    "input": 131072,
    "output": 65536
  },
  "modalities": {
    "input": [
      "text",
      "image",
      "audio",
      "video"
    ],
    "output": [
      "text",
      "audio"
    ]
  },
  "reasoning": {
    "supported": true,
    "efforts": []
  },
  "pricing": null,
  "features": [
    "tools"
  ],
  "info": {
    "status": "active",
    "release_date": "2026-09-15",
    "knowledge_cutoff": null,
    "description": "Default low-latency voice agent model for real-time dialogue without reasoning delays.",
    "docs": "https://ai.google.dev/gemini-api/docs/models/gemini-3.8-live",
    "verified_at": "2026-10-11"
  }
}

官方来源