DEDUPED Hot AI repos. Real use cases. What people build. Get the Monday 5

How to speed up local LLM prompt processing on an RTX 30 card with GPU Unlock

Install the gpu_unlock module so FP16 and BF16 tensor math runs at full rate for vLLM, PyTorch and Transformers.

Short answer: Run git clone https://github.com/Zagitta/gpu-unlock, cd gpu-unlock and make install; then Append gpu_unlock.targets=all to your kernel command line (GRUB, systemd-boot or UKI, as the README shows); on Debian or Ubuntu run sudo update-grub, then reboot. All steps.

system-change. This installs a kernel module that loads before the NVIDIA driver; keep a known good boot entry in case the system does not start.

unofficial. This is not supported by NVIDIA. The README says harm is very unlikely but the card may run hotter.

Who it is forA home lab user serving unquantized or AWQ models on an RTX 3060 to 3090 who waits on long prompts.

The problemGeForce cards run tensor core math with FP32 accumulation at half the rate of the same silicon sold as a pro card.

The outcomeFaster prefill and time to first token on FP16, BF16, AWQ or GPTQ models, with bit-identical outputs.

What you need first

  • Linux (Windows is not supported; the README points to a third party port)
  • An Ampere RTX 30 card (stable) or an Ada RTX 40 or Turing RTX 20 card (experimental); RTX 50 is not supported
  • Root access to install a kernel module and edit the kernel command line

Steps

  1. Run git clone https://github.com/Zagitta/gpu-unlock, cd gpu-unlock and make install
  2. Append gpu_unlock.targets=all to your kernel command line (GRUB, systemd-boot or UKI, as the README shows)
  3. On Debian or Ubuntu run sudo update-grub, then reboot
  4. Rerun your usual vLLM or Transformers benchmark and compare prefill speed

Prompt for your agent

Walk me through installing Zagitta/gpu-unlock on this <distro> machine with an <RTX card>: build the module, add gpu_unlock.targets=all to the kernel command line for my bootloader, and give me a before and after benchmark to run.

Caveat. This card is written from the Zagitta/gpu-unlock README, not from a test run by Deduped. Before you start, make sure you have Linux. Check the repo for newer instructions before you rely on these steps.

Use this in your AI agent

Ask your agent

Using the Deduped MCP, get the setup steps for Zagitta/gpu-unlock and "Speed up local LLM prompt processing on an RTX 30 card"

Add Deduped to Claude Code once

claude mcp add --transport http deduped https://dedupedai.com/mcp

Cursor, ChatGPT, others

Share: X LinkedIn Hacker News Email Copy link

Third-party repos we don't own. AI-written use cases can be wrong. Disclaimer