Quick post about a change I made that’s worked out well.
I was using OpenAI API for automations in n8n — email summaries, content drafts, that kind of thing. Was spending ~$40/month.
Switched everything to Ollama running locally. The migration was pretty straightforward since n8n just hits an HTTP endpoint. Changed the URL from api.openai.com to localhost:11434 and updated the request format.
For most tasks (summarization, classification, drafting) the local models are good enough. Complex reasoning is worse but I don’t need that for automation workflows.
Hardware: i7 with 16GB RAM, running Llama 3 8B. Plenty fast for async tasks.
I only ever use my local ai for home assistant voice assistant on my phone, but it’s more of a gimmick/party trick since I only have temperatures sensors currently (only got into ha recently) and it can’t access WiFi so it’s just quietly sitting unloaded on my truenas server
Running any LLM on truenas is not awesome. I’ve tried it with GPU passthrough and it’s just too much overhead. I may just burn all my stuff down and restart with Proxmox, run Truenas core inside just for NAS. The idea of a converged nas+virtualization is wonderful, but it’s just not there.
The host networking model alone is such a pain, then you get into performance stuff. I still like Truenas a lot, but I think that Proxmox is probably still the better platform.