33
0

OpenClaw 配置

2026-08-04
2026-08-05
OpenClaw

配置openclaw模型

一、打开终端

输入openclaw config 进入配置选项

openclaw config

二、选择Local (this machine)

◆  Where will the Gateway run?
│  ● Local (this machine) (Gateway reachable (ws://127.0.0.1:18789))
│  ○ Remote (info-only)
└

三、配置模型

◆  What do you want to configure?
│  ○ Workspace
│  ● Model (Pick provider + credentials)
│  ○ Web tools
│  ○ Gateway
│  ○ Daemon
│  ○ Channels
│  ○ Plugins
│  ○ Skills
│  ○ Health check
│  ○ Done
└

四、模型选择

此次模型选择DeepSeek，也可以根据需要选择其他模型

如果是1Panel面板已配置DeepSeek模型，可以进入工作目录修改openclaw.json 配置文件，删除DeepSeek部分，并重新进入终端下载DeepSeek插件，重新生成配置文件，这样更准确

◆  Model/auth provider
│
│  Search:
│  ○ OpenAI
│  ○ Anthropic
│  ○ xAI (Grok)
│  ○ Google
│  ○ OpenRouter
│  ○ Arcee AI
│  ○ BytePlus
│  ○ Cerebras
│  ○ Chutes
│  ○ Cloudflare AI Gateway
│  ○ Codex
│  ○ Cohere
│  ○ Copilot
│  ○ Custom Provider
│  ○ DeepInfra
│  ● DeepSeek (API key)
│  ○ Fireworks
│  ...
│  ↑/↓ to select • Enter: confirm • Type: to search
└

五、选择DeepSeek插件下载方式

◆  Install DeepSeek plugin?
│  ○ Download from ClawHub (clawhub:@openclaw/deepseek-provider)
│  ● Download from npm (@openclaw/deepseek-provider)
│  ○ Skip for now
└

六、输入模型的key

模型key去模型官网获取，此次DeepSeek官网获取https://platform.deepseek.com/

◆  Enter DeepSeek API key
│  _
└

七、选择使用的模型，空格选择，回车确认

◆  Models in /model picker (multi-select)
│
│  Search: 
│  ◼ deepseek/deepseek-v4-flash (DeepSeek V4 Flash · ctx 977k · reasoning · 
alias: 
│  DeepSeek)
│  ◼ deepseek/deepseek-v4-pro
│  ◻ deepseek/deepseek-chat
│  ◻ deepseek/deepseek-reasoner
│  ↑/↓ to navigate • Space/Tab: select • Enter: confirm • Type: to search
└

DeepSeek插件生成的参数，可以参考

"providers": {
      "deepseek": {
        "api": "openai-completions",
        "baseUrl": "https://api.deepseek.com",
        "models": [
          {
            "api": "openai-completions",
            "compat": {
              "maxTokensField": "max_tokens",
              "supportsReasoningEffort": true,
              "supportsUsageInStreaming": true
            },
            "contextWindow": 1000000,
            "cost": {
              "cacheRead": 0.028,
              "cacheWrite": 0,
              "input": 0.14,
              "output": 0.28
            },
            "id": "deepseek-v4-flash",
            "input": [
              "text"
            ],
            "maxTokens": 384000,
            "name": "DeepSeek V4 Flash",
            "reasoning": true
          },
          {
            "api": "openai-completions",
            "compat": {
              "maxTokensField": "max_tokens",
              "supportsReasoningEffort": true,
              "supportsUsageInStreaming": true
            },
            "contextWindow": 1000000,
            "cost": {
              "cacheRead": 0.145,
              "cacheWrite": 0,
              "input": 1.74,
              "output": 3.48
            },
            "id": "deepseek-v4-pro",
            "input": [
              "text"
            ],
            "maxTokens": 384000,
            "name": "DeepSeek V4 Pro",
            "reasoning": true
          },
          {
            "api": "openai-completions",
            "compat": {
              "maxTokensField": "max_tokens",
              "supportsUsageInStreaming": true
            },
            "contextWindow": 131072,
            "cost": {
              "cacheRead": 0.028,
              "cacheWrite": 0,
              "input": 0.28,
              "output": 0.42
            },
            "id": "deepseek-chat",
            "input": [
              "text"
            ],
            "maxTokens": 8192,
            "name": "DeepSeek Chat",
            "reasoning": false
          },
          {
            "api": "openai-completions",
            "compat": {
              "maxTokensField": "max_tokens",
              "supportsReasoningEffort": false,
              "supportsUsageInStreaming": true
            },
            "contextWindow": 131072,
            "cost": {
              "cacheRead": 0.028,
              "cacheWrite": 0,
              "input": 0.28,
              "output": 0.42
            },
            "id": "deepseek-reasoner",
            "input": [
              "text"
            ],
            "maxTokens": 65536,
            "name": "DeepSeek Reasoner",
            "reasoning": true
          }
        ]
      }
    }






配置web 搜索工具

一、打开终端

输入openclaw config 进入配置选项

openclaw config

二、选择Local (this machine)

◆  Where will the Gateway run?
│  ● Local (this machine) (Gateway reachable (ws://127.0.0.1:18789))
│  ○ Remote (info-only)
└

三、配置Web tool

◆  What do you want to configure?
│  ○ Workspace
│  ○ Model
│  ● Web tools (Configure web search (Perplexity/Brave) + fetch)
│  ○ Gateway
│  ○ Daemon
│  ○ Channels
│  ○ Plugins
│  ○ Skills
│  ○ Health check
│  ○ Skip for now
└
◆  Enable web_search?
│  ● Yes / ○ No
└

五、选择工具

此次选择Tavily Search，也可以根据需要选择其他工具

◆  Search provider
│
│  Search:
│  ○ Brave Search
│  ○ DuckDuckGo Search (experimental)
│  ○ Exa Search
│  ○ Firecrawl Search
│  ○ Gemini (Google Search)
│  ○ Grok (xAI)
│  ○ Kimi (Moonshot)
│  ○ MiniMax Search
│  ○ Ollama Web Search
│  ○ Parallel Search
│  ○ Parallel Search (Free)
│  ○ Perplexity Search
│  ○ SearXNG Search
│  ● Tavily Search (Structured results with domain filters and AI answer 
│  summaries)
│  ○ Skip for now
│  ↑/↓ to select • Enter: confirm • Type: to search
└

六、输入key

key去Tavily官网获取https://www.tavily.com/

◆  Tavily API key
│  _
└

七、开启web_fetch

◆  Enable web_fetch (keyless HTTP fetch)?
│  ● Yes / ○ No
└






其他配置

输入openclaw config 后根据需要配置

◆  What do you want to configure?
│  ● Workspace (Set workspace + sessions)
│  ○ Model
│  ○ Web tools
│  ○ Gateway
│  ○ Daemon
│  ○ Channels
│  ○ Plugins
│  ○ Skills
│  ○ Health check
│  ○ Done
└






支持与分享

如果这篇文章对你有帮助，欢迎分享给更多人或者给予支持！

OpenClaw 配置
/archives/019fd0f6-9891-75e1-9c59-5ff0a005168b
作者
北祢
发布于
2026-08-04
许可协议
CC BY-NC-SA 4.0