adingbatponder  πŸ‘ΎMay 21, 2025
Is there a good resource for knowing which #gpu and #motherboard combinations (hardware and #bios) should work for #pciepassthrough #iommu under #proxmox for #llm ?
adingbatponder  πŸ‘ΎMay 20, 2025
#llm #ollama #Llama3 #nixos #proxmox #openwebui #nvidia #Nvidia #a4000 #GPU #tailnet #tailscale #pciepassthrough #HomeLab #SelfHost #SelfHosting #AI #vm #NixOS
https://youtu.be/9hni6rLfMTg
Build a private, self-hosted LLM server with Proxmox, PCle passthrough, Ollama, Open WebUI & NixOS

YouTube
Maddy, Friendly Local CryptidJan 27, 2025

So heads up for anyone looking at snagging a deal on the X670E Tomahawk for Virtualization - there's a single IOMMU group shared between the bottom PCIe slot (Physical 16x, 2 lanes) and two bottom M.2 slots, as well as what seems to be part of the chipset (I lost front panel USB while fucking around and finding out, trying to assign devices in the same group to a VM)

Not ideal, but I can still work with this.

Thankfully the top two physical 16x slots and the top two M.2 slots are all on separate IOMMU groups!

#Homelab #Virtualization #PCIePassthrough