Post

Lemonade Server

Lemonade Server is an open-source local AI runtime with OpenAI-compatible APIs for chat, vision, image generation, transcription, and speech. It auto-detects GPU and NPU hardware and provides a web control panel for model management.

Lemonade Server

Installation

Default install:

1
bash -c "$(curl -fsSL https://raw.githubusercontent.com/community-scripts/ProxmoxVE/main/ct/lemonade-server.sh)"
CPU: 4 cores RAM: 8192 MB Disk: 80 GB OS: Debian 13

Notes

Remote access is configured with host=0.0.0.0, which also exposes the internal control endpoints (/internal/*, including shutdown and config). Secure it before use: run systemctl edit lemond.service, add [Service] followed by Environment="LEMONADE_API_KEY="</code>, then systemctl restart lemond. LEMONADE_API_KEY protects every endpoint; LEMONADE_ADMIN_API_KEY alone only protects /internal/* and leaves /api, /v0 and /v1 open.</div> </div>
Models download on demand — increase disk size if you plan to run large models.
## Web Interface
Port: 13305
## Links - [Official Website](https://lemonade-server.ai/) - [Documentation](https://lemonade-server.ai/docs/) ---
Source code
Loading...