Setup OpenClaw with LM Studio

Setup LM Studio and turn on the server option.

Next edit openclaw.json that should be in ~/.openclaw

Add this block with your details:

"models": {
    "mode": "merge",
    "providers": {
      "lmstudio": {
        "baseUrl": "BASE_URL",
        "apiKey": "lmstudio",
        "api": "openai-responses",
        "models": [
          {
            "id": "model_id",
            "name": "model_name",
            "reasoning": false,
            "input": [
              "text"
            ],
            "cost": {
              "input": 0,
              "output": 0,
              "cacheRead": 0,
              "cacheWrite": 0
            },
            "contextWindow": 131072,
            "maxTokens": 4096
          }
        ]
      }
    }
  }

The base url will be the same as LM Studio but with v1 appended.

For me it’s local so that’s http://127.0.0.1:1234/v1

My model id is: qwen3-30b-a3b-thinking-2507

(use this for model id (model_id) and name (model_name)

Go to the agents section in your json and change the primary model and add your model: 

"model": {
   "primary": "lmstudio/qwen3-30b-a3b-thinking-2507"
},
"models": {
   "lmstudio/qwen3-30b-a3b-thinking-2507": {}
}

Restart OpenClaw and onboard.

Choose quickstart and make sure it detects your model.

Skip all the steps and use existing values.

Once this is done setup is done and you should be ready to go!