Re:Show HN: Llama 3.1 70B on a single RTX 3090 via NVMe-to-G...
by Aurornis on Sunday February 22, 2026 @01:03AM
(#47108644)
Cool project. Can you provide more details about your DKMS patching process for consumer GPUs? This would be fun to try out, but I’d need some more details on that patch process first.
[ Parent
| on HN ]
Re:Show HN: Llama 3.1 70B on a single RTX 3090 via NVMe-to-G...
by Maxious on Sunday February 22, 2026 @03:58AM
(#47109499)
the nvidia open source driver has been modded previously to unlock enterprise paywalled features like p2p gpu comms https://blog.chlc.cc/p/rtx4090-p2p-unlocked and vGPU splitting https://open-iov.org/index.php/VGPU_Unlock