openclaw.json配置示例

基于免费的nvidia的大模型配置的,记得替换json里的apiKey。 apikey 可以在这里注册获取https://build.nvidia.com/models?filters=nimType%3Anim_type_preview\&orderBy=weightPopular%3ADESC

复制代码
{
  "agents": {
    "defaults": {
      "workspace": "/Users/xy/.openclaw/workspace",
      "model": {
        "primary": "nvidia/moonshotai/kimi-k2-instruct"
      },
      "models": {
        "nvidia/moonshotai/kimi-k2-instruct": {}
      }
    }
  },
  "gateway": {
    "mode": "local",
    "auth": {
      "mode": "token",
      "token": "openclaw gateway run"
    },
    "port": 18789,
    "bind": "loopback",
    "tailscale": {
      "mode": "off",
      "resetOnExit": false
    },
    "controlUi": {
      "allowInsecureAuth": true
    },
    "nodes": {
      "denyCommands": [
        "camera.snap",
        "camera.clip",
        "screen.record",
        "contacts.add",
        "calendar.add",
        "reminders.add",
        "sms.send",
        "sms.search"
      ]
    }
  },
  "session": {
    "dmScope": "per-channel-peer"
  },
  "tools": {
    "profile": "coding",
    "web": {
      "search": {
        "provider": "duckduckgo",
        "enabled": true
      }
    }
  },
  "auth": {
    "profiles": {
      "moonshotai:default": {
        "provider": "moonshotai",
        "mode": "oauth"
      }
    }
  },
  "plugins": {
    "entries": {
      "duckduckgo": {
        "enabled": true
      }
    }
  },
  "wizard": {
    "lastRunAt": "2026-04-02T09:19:17.192Z",
    "lastRunVersion": "2026.4.1",
    "lastRunCommand": "onboard",
    "lastRunMode": "local"
  },
  "meta": {
    "lastTouchedVersion": "2026.4.1",
    "lastTouchedAt": "2026-04-02T09:19:17.241Z"
  },
  "models": {
    "providers": {
      "nvidia": {
        "baseUrl": "https://integrate.api.nvidia.com/v1",
        "apiKey": "nvapi-你nvidia的token,在这里注册获取https://build.nvidia.com/models?filters=nimType%3Anim_type_preview&orderBy=weightPopular%3ADESC",
        "auth": "api-key",
        "api": "openai-completions",
        "headers": {},
        "models": [
          {
            "id": "moonshotai/kimi-k2-instruct",
            "name": "moonshotai/kimi-k2-instruct",
            "reasoning": false,
            "input": [
              "text"
            ],
            "cost": {
              "input": 0,
              "output": 0,
              "cacheRead": 0,
              "cacheWrite": 0
            },
            "contextWindow": 128000,
            "maxTokens": 8192
          }
        ]
      }
    }
  }
}
相关推荐
无心水21 小时前
【Hermes:安全、权限与生产环境】38、Hermes Agent 安全四层纵深:最小权限原则从理论到落地的完全指南
人工智能·安全·mcp协议·openclaw·养龙虾·hermes·honcho
原来是猿21 小时前
网络计算器:理解序列化与反序列化(中)
linux·运维·服务器·网络·tcp/ip
c1s2d3n4cs1 天前
Qt模仿nlohmann::json进行序列化和反序列化
开发语言·qt·json
一只很酸de橘子1 天前
PostgreSQL 生成 JSON 字符串
postgresql·json
m0_738120721 天前
ctfshow靶场SSRF部分——基础绕过到协议攻击解题思路与技巧(一)
服务器·前端·网络·安全·php
2301_780789661 天前
“数字珍珠港”再现:西北能源基地DNS篡改事件深度复盘与防护升级
运维·服务器·网络·tcp/ip·网络安全·智能路由器·能源
无心水1 天前
【Hermes:安全、权限与生产环境】39、智能体也会犯错?Hermes 纠错、回滚与遗忘机制全指南 —— 让 AI 的错误像 Git 一样可逆可控
人工智能·git·安全·mcp协议·openclaw·hermes·honcho
南境十里·墨染春水1 天前
linux学习进展 守护进程
linux·服务器·学习
AI自动化工坊1 天前
Hermes Agent 日处理 224B tokens:自改进循环与 Kanban 任务板架构深度解析
架构·ai agent·openclaw·hermes agent