DEDUPED Hot AI repos, turned into use cases Get the Monday 5
802 stars · C

How to use GPU Unlock Ada: what you can do with it

What is GPU Unlock Ada? GPU Unlock for Ada is a Linux kernel module that runs at boot, before the stock NVIDIA driver loads, and switches supported RTX 40 cards to full-rate FP16 and BF16 tensor math with FP32 accumulation. The authors measured about 2x in instruction benchmarks on an RTX 4090 (165 to 331 TFLOPS) and smaller gains in real GEMM workloads.

How do I use it? Run git clone https://github.com/JustAResearcher/gpu-unlock-ada.git and cd gpu-unlock-ada; then run make install to build the package locally, which downloads and verifies the pinned NVIDIA inputs and generates the firmware on your machine; Find the card's full PCI address with lspci -D and add gpu_unlock.targets= with that exact address to the kernel command line. Full steps below.

Trended earlier

View repo on GitHub

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

What do you want to do?

3 use cases

  • Unlock full-rate FP16 and BF16 tensor math on an RTX 4090

    Build the kernel module, add one boot option, cold boot, and check the kernel log for the unlock.

    View prompt
    On my Linux machine, build and install JustAResearcher/gpu-unlock-ada with `make install`, add gpu_unlock.targets=<my RTX 4090 PCI address> to the kernel command line, keep a stock boot entry, and tell me when to cold boot. After boot, report the gpu_unlock lines from the kernel log and whether `gpu_unlock: unlocked` appears for that card.

  • Benchmark stock vs unlocked tensor throughput

    Run matched stock and unlocked measurements so you know what the unlock does for your card and workload.

    View prompt
    Using JustAResearcher/gpu-unlock-ada on <my GPU>, run the diagnostics/bench instruction benchmark and <my workload> in stock mode, then again after a cold boot with the unlock enabled, with the same kernel, driver and power cap. Report a before and after table and confirm the kernel log shows `gpu_unlock: unlocked`.

  • Test the experimental RTX 4070 Ti SUPER or 4060 Ti path

    Opt in to the experimental AD103 or AD106 target with its flag and one exact PCI address.

    View prompt
    Install JustAResearcher/gpu-unlock-ada and enable the experimental path for my <RTX 4070 Ti SUPER or RTX 4060 Ti 16GB> at <PCI address> with its experimental flag, keeping a stock boot entry. After a cold boot, report the gpu_unlock kernel log lines and whether the card unlocked.

Quick start with GPU Unlock Ada

  1. Check you have what it needs: A Linux machine you fully control, with a spare stock boot entry to fall back to; RTX 4090 (10de:2684); RTX 4070 Ti SUPER (10de:2705) and RTX 4060 Ti 16GB (10de:2805) are experimental and validated on one card each; Kernel headers and a working DKMS toolchain, with CONFIG_CRYPTO_LIB_SHA256 in the kernel.
  2. Open JustAResearcher/gpu-unlock-ada on GitHub and read the README and license.
  3. Run git clone https://github.com/JustAResearcher/gpu-unlock-ada.git and cd gpu-unlock-ada
  4. Run make install to build the package locally, which downloads and verifies the pinned NVIDIA inputs and generates the firmware on your machine
  5. Find the card's full PCI address with lspci -D and add gpu_unlock.targets= with that exact address to the kernel command line
  6. Power the machine off and start it again for the first attempt, since a warm reboot may not give the card a fresh state

These steps come from our top use case, Unlock full-rate FP16 and BF16 tensor math on an RTX 4090, which also has a copy-paste agent prompt.

Use this in your AI agent

Ask your agent

Using the Deduped MCP, get the setup steps for JustAResearcher/gpu-unlock-ada

Add Deduped to Claude Code once

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

Cursor, ChatGPT, others

GPU Unlock Ada alternatives · Best AI repos for Software and DevOps

About this repo

  • Who it is forLinux users who train or run models on their own RTX 4090, and tinkerers willing to test experimental support on an RTX 4070 Ti SUPER or RTX 4060 Ti 16GB.
  • Why it is hot802 stars, launched 3 days ago, talked about in 2 AI sources we track
  • LanguageC
  • Sourcegithub.com/JustAResearcher/gpu-unlock-ada · Back to Hot repos

Requirements

  • A Linux machine you fully control, with a spare stock boot entry to fall back to
  • RTX 4090 (10de:2684); RTX 4070 Ti SUPER (10de:2705) and RTX 4060 Ti 16GB (10de:2805) are experimental and validated on one card each
  • Kernel headers and a working DKMS toolchain, with CONFIG_CRYPTO_LIB_SHA256 in the kernel
  • The stock NVIDIA driver
  • Secure Boot turned off, or the out-of-tree module signed with your own key
  • Your own stock benchmark numbers taken before you enable it

Add this badge to your README

Maintain GPU Unlock Ada? Show readers where to find these use cases. Use cases on Deduped

Markdown 

[![Use cases on Deduped](https://dedupedai.com/badge/JustAResearcher/gpu-unlock-ada.svg)](https://dedupedai.com/use-cases/justaresearcher-gpu-unlock-ada/?utm_source=github&utm_medium=badge&utm_campaign=JustAResearcher-gpu-unlock-ada)

HTML 

<a href="https://dedupedai.com/use-cases/justaresearcher-gpu-unlock-ada/?utm_source=github&amp;utm_medium=badge&amp;utm_campaign=JustAResearcher-gpu-unlock-ada"><img src="https://dedupedai.com/badge/JustAResearcher/gpu-unlock-ada.svg" alt="Use cases on Deduped" height="20"></a>

A static image, no tracking script. About the badge

Questions about GPU Unlock Ada

What is GPU Unlock Ada?

GPU Unlock for Ada is a Linux kernel module that runs at boot, before the stock NVIDIA driver loads, and switches supported RTX 40 cards to full-rate FP16 and BF16 tensor math with FP32 accumulation. The authors measured about 2x in instruction benchmarks on an RTX 4090 (165 to 331 TFLOPS) and smaller gains in real GEMM workloads. It is an open-source project on GitHub: https://github.com/JustAResearcher/gpu-unlock-ada.

How do I use GPU Unlock Ada?

Run git clone https://github.com/JustAResearcher/gpu-unlock-ada.git and cd gpu-unlock-ada; then run make install to build the package locally, which downloads and verifies the pinned NVIDIA inputs and generates the firmware on your machine; Find the card's full PCI address with lspci -D and add gpu_unlock.targets= with that exact address to the kernel command line. The use cases on this page give the full steps and a copy-paste agent prompt for each job.

Who is GPU Unlock Ada for?

Linux users who train or run models on their own RTX 4090, and tinkerers willing to test experimental support on an RTX 4070 Ti SUPER or RTX 4060 Ti 16GB.

What can I do with GPU Unlock Ada?

3 use cases so far, for example: Unlock full-rate FP16 and BF16 tensor math on an RTX 4090; Benchmark stock vs unlocked tensor throughput; Test the experimental RTX 4070 Ti SUPER or 4060 Ti path. Each has steps and a copy-paste agent prompt.

What do I need to run GPU Unlock Ada?

A Linux machine you fully control, with a spare stock boot entry to fall back to; RTX 4090 (10de:2684); RTX 4070 Ti SUPER (10de:2705) and RTX 4060 Ti 16GB (10de:2805) are experimental and validated on one card each; Kernel headers and a working DKMS toolchain, with CONFIG_CRYPTO_LIB_SHA256 in the kernel; The stock NVIDIA driver.

What are alternatives to GPU Unlock Ada?

Other hot AI repos in Software and DevOps: Deep Infer, Harness2, Orca, Opencode, Graphify.

Is GPU Unlock Ada safe to run?

Deduped does not own, audit or vouch for it. Read the README and license, run it on your own systems first and review every command before you run it.