Run OpenClaw on Proxmox VE

Proxmox VE is a favorite among homelabbers and self-hosters, and it's an excellent platform for running OpenClaw. You have two main deployment options: LXC containers (lightweight, efficient) or full virtual machines (more isolation, GPU passthrough support).

For most users, an LXC container running Docker is the sweet spot. Create a Debian or Ubuntu LXC, install Docker inside it, and deploy your chosen OpenClaw wrapper using docker-compose. This gives you near-native performance with minimal overhead.

If you want to run local AI models alongside OpenClaw (via Ollama), you'll need GPU passthrough — which requires a full VM in Proxmox. Create a VM, pass through your NVIDIA or AMD GPU, install Docker, and run both OpenClaw and Ollama in the same environment.

The Proxmox community has created several one-click helper scripts (like those from tteck) that automate the entire setup process. These scripts create an LXC container, install Docker, and configure networking in under 2 minutes.

Resource allocation depends on your use case. A basic OpenClaw installation works well with 2 CPU cores and 2 GB of RAM. If running local models, allocate at least 4 cores, 16 GB RAM, and a dedicated GPU.