LM Studio (Local)
LM Studio is a desktop application for running local LLMs with a user-friendly interface. It provides an OpenAI-compatible API server, making it a good option for users who prefer a GUI for model management.
Modes
The extension supports LM Studio in two ways:
1. HTTP Mode (Recommended)
Connects directly to the LM Studio local server API.
Setup:
Download and install LM Studio from lmstudio.ai.
Download a model within LM Studio.
Start the local server: Developer tab → Start Server.
Configure in Burp: open Burp AI Agent → Settings → AI Backend and set:
Preferred Backend
LM Studio (select from dropdown)
LM Studio URL
http://127.0.0.1:1234 (default)
LM Studio Model
The identifier of the model you loaded (check LM Studio's server log)
2. Auto-Start Mode
The extension can launch the LM Studio server for you.
Setup: In Settings → AI Backend, set:
LM Studio Auto-Start
ON
LM Studio Server Command
The command that starts the LM Studio server
When Auto-Start is enabled, the extension runs the server command automatically when this backend is selected.
Configuration
LM Studio URL
http://127.0.0.1:1234
Base URL for the LM Studio server API.
LM Studio Model
lmstudio
Model identifier to use.
Auto-Start
On
Automatically start the LM Studio server.
Server Command
lms server start
Custom command to launch the server.
Timeout
120 seconds
Request timeout (range: 30–3600 seconds). Increase for larger models or complex analyses.
Troubleshooting
Connection refused: Ensure the LM Studio server is running and the port matches your configuration.
Model not found: Verify the model identifier matches the loaded model in LM Studio.
Timeout errors: Increase the Timeout setting for large responses. Consider using a smaller model if timeouts persist.
Slow responses: LM Studio performance depends on your hardware. GPU inference is significantly faster than CPU-only.
Last updated