Guide: LibreChat + LM Studio on Windows 11

[mrcodetastic]

I have a windows PC in a remote location that runs LM Studio and various models. Instead of using the LM Studio API from my local device, I simply want the remote PC to run the model using LM Studio and host the web chat interface as well.

LM Studio was installed as usual, with the Local Server enabled and a Model Loaded (in this case 'qwen3.5-4b').

LibreChat was installed on the same windows PC using docker, using the standard 3 or so steps: https://www.librechat.ai/docs/local/docker

However, before running the docker image I needed to:

1. Create a 'docker-compose.override.yml' which contains

services:
   api:
     volumes:
     - type: bind
       source: ./librechat.yaml
       target: /app/librechat.yaml
     image: registry.librechat.ai/danny-avila/librechat:latest

2. Create 'librechat.yaml' based off the example file and add the following under the endpoints: section

  custom:
    - name: "LM Studio"
      apiKey: "sk-1234"
      baseURL: "http://host.docker.internal:1234/v1"
      models: 
        default:
            - qwen3.5-4b    # Change to the model running on LM Studio.
        fetch: true
      titleConvo: true
      titleMethod: "completion"
      titleModel: "current_model"
      summarize: false
      summaryModel: "current_model"
      forcePrompt: false
      modelDisplayLabel: "Local Chat"
      dropParams: ["stop", "user", "frequency_penalty", "presence_penalty"]

Note: When a Docker container (such as the one running LibreChat) needs to connect to a service on the host machine, you must use host.docker.internal. Using 127.0.0.1 will not work because Docker containers run in isolated environments—127.0.0.1 refers to the container itself, not the host system.

Thanks to this previous discussion for helping: #1836