Skip to main content
LM Studio provides a desktop GUI for downloading, running, and serving local models. It includes a built-in OpenAI-compatible server.

​
Quick Start

  1. Download and install LM Studio
  2. Load a model in the GUI
  3. Start the local server: Settings > Local Server > Start

​
Config

{
  "providers": {
    "lmstudio": {
      "baseUrl": "http://localhost:1234/v1",
      "api": "openai-completions",
      "models": [
        {
          "id": "qwen2.5-coder-32b-instruct",
          "name": "Qwen 2.5 Coder 32B",
          "contextWindow": 131072,
          "maxTokens": 8192,
          "cost": { "input": 0, "output": 0, "cacheRead": 0, "cacheWrite": 0 }
        }
      ]
    }
  }
}

​
Use it

{
  "agents": [
    { "name": "local-coder", "model": "lmstudio:qwen2.5-coder-32b-instruct" }
  ]
}

​
Auto-Discovery

polpo models scan
Scans localhost:1234 for a running LM Studio server.

​
Provider Details

Provider IDlmstudio (custom)
Default port1234
API typeopenai-completions
Base URLhttp://localhost:1234/v1
API keyNot required
CostFree (runs locally)

​
Notes

  • LM Studio’s model IDs may differ from Ollama’s. Check what ID LM Studio reports in its server logs.
  • LM Studio has a user-friendly GUI for model management — good for users who prefer not to use the command line.
  • Supports both CPU and GPU inference. GPU is strongly recommended for usable performance.