3090 unlocked: +80% AI compute via firmware bug, 4090 nearly doubles
lxfater · x · 2026-10-24
The open-source gpu-unlock project exploits a GPU boot-firmware vulnerability to write speed-selector registers before the NVIDIA driver binds the card, unlocking the FP16/BF16 instruction issue rate capped on consumer cards. A 3090 jumped from 75 to 135 TFLOPS in PyTorch tests at the same 360W limit (+80%), and a 4090 branch nearly doubled instruction throughput from 165 to 331 TFLOPS. It's a Linux kernel module combining the V67/PLM technique with pre-driver timing, so no driver patching is needed.
More from Infra
- Used NVIDIA DGX Spark reselling at $6,200, well above its ~$4,000 retail price — AIFlow_ML · 2026-10-25
- SGLang Matches DeepInfer on Prefill, Adds 7.4% Decode Throughput on 64×GB300 — BanghuaZ · 2026-10-25
- Poldi: A Fully Offline Local Voice AI Assistant That Runs on 8GB VRAM — rhinodevil · 2026-10-25
- kalomaze says he no longer has meaningful GPU access, back to toy baselines — kalomaze · 2026-10-25
- Is the explosion of LLM inference engines healthy competition or fragmentation? — Anstellos · 2026-10-25
- Jev ecosystem explodes: 5 open-source apps from a 7.1s flight-search agent to SQL-in-English — udmrzn · 2026-10-25