Lemonade
Lemonade Server lets you run local LLMs on your PC’s NPU and GPU. Free, private, and OpenAI-compatible.
Prerequisites
Section titled “Prerequisites”- Install Lemonade Server - Download from lemonade-server.ai
- Install a model - Use the Lemonade GUI or CLI to download a model
Setup in Multi
Section titled “Setup in Multi”Setting up Lemonade in Multi
- Open the Multi panel → Settings (gear icon)
- Click Add Profile
- Select Lemonade as the provider
- Set the base URL (default:
http://localhost:13305/api/v1) - Select your model
- Save
Configuration Options
Section titled “Configuration Options”| Option | Description |
|---|---|
| Base URL | Lemonade server URL (default: http://localhost:13305/api/v1) |
| Model ID | The model to use (e.g., Llama-3.2-1B-Instruct-Hybrid) |
Free - Lemonade Server is open-source and runs on your own hardware.