LM Studio Local Setup

A complete setup guide and example scripts for running LM Studio locally with Python SDK integration.

Overview

This project provides:

• Step-by-step LM Studio installation and configuration
• Python examples using both LM Studio SDK and OpenAI-compatible API
• Connection testing utilities
• Sample chat completion implementations

Prerequisites

• Python 3.13 or higher
• Windows, macOS, or Linux
• At least 8GB RAM (16GB+ recommended for larger models)

LM Studio Installation

1. Download LM Studio

   • Visit LM Studio official website
   • Download the installer for your operating system
   • Install following the standard installation process

2. Download a Model

   • Launch LM Studio
   • Go to the "Discover" tab
   • Search for and download a model

3. Start the Local Server

   • Navigate to the "Developer" tab in LM Studio
   • Toggle "Status: Stopped" to "Status: Running"

Python Environment Setup

Using uv (Recommended)

uv is a fast Python package installer and resolver that works great with this project.

1. Install uv

   curl -LsSf https://astral.sh/uv/install.sh | sh

2. Clone this repository

   git clone https://github.com/medvoice-research/lmstudio-local-setup.git
   cd lmstudio-local-setup

3. Install dependencies with uv

   uv sync

Using Traditional pip/venv

1. Create a virtual environment

   python -m venv venv
   
   # On Windows
   venv\Scripts\activate
   
   # On macOS/Linux
   source venv/bin/activate

2. Install dependencies with pip

   pip install -r requirements.txt

Configuration

1. Environment Variables (Optional for local testing) Create a .env file in the project root:

   cp .env.example .env

   LMSTUDIO_SERVER_HOST=localhost:1234
   LMSTUDIO_API_HOST=http://localhost:1234/v1
   LMSTUDIO_API_KEY=lm-studio
   MODEL_CHOICE=your-preferred-model-name

Usage Examples

1. Test Connection

Run the comprehensive connection test:

python test_lmstudio_connection.py

This script will:

• Test LM Studio SDK connection
• Test OpenAI-compatible API connection
• Perform a sample chat completion
• Provide troubleshooting information if connections fail

2. Simple Hello World

Run the basic example:

python lmstudio_hello_world.py

3. Using OpenAI-Compatible API

from openai import OpenAI

client = OpenAI(
    base_url="http://localhost:1234/v1",
    api_key="lm-studio"
)

response = client.chat.completions.create(
    model="your-model-name",
    messages=[
        {"role": "user", "content": "Hello, how are you?"}
    ]
)

print(response.choices[0].message.content)

Troubleshooting

See docs/troubleshooting.md for information on increasing VRAM on Apple Silicon.

License

See LICENSE file for details.

Additional Resources

• LM Studio Documentation
• LM Studio Python SDK
• OpenAI Python Library