What happened?
A developer from the XDA community ran an unusual experiment: he gave a local language model (LLM) full control over his Proxmox node. The outcome exceeded all expectations — the AI successfully managed virtual machines, containers, and networks.
How it was done?
The experiment used the Llama 3.1 (70B) model, deployed on a separate server. A Python interpreter was connected via the model's API, allowing it to execute commands in Proxmox through its REST API. As a safety measure, all actions were logged, and critical operations required confirmation.
Key highlights:
- The model independently created and deleted LXC containers.
- Configured network bridges and VLANs.
- Managed backups and VM migrations.
- In one test, the AI itself detected an unused disk and suggested freeing it.
Conclusion
The experiment shows that modern local LLMs can not only generate text but also perform complex administrative tasks. However, full trust in AI assistants is still premature — human oversight and clear boundaries are necessary.
This material is for informational purposes only and does not constitute investment advice.
A virtual card in 2 minutes
Pay for subscriptions, AI tools, travel, and international stores. Top up via USDT-TRC20 with no acquiring fees.
