xanpavle/rocmfix

Auto-detect AMD GPUs on Windows/Linux and instantly get the correct HSA_OVERRIDE_GFX_VERSION for local AI (ROCm/HIP).

Python

9 commits

updated Sep 19, 2026

See the code

See what people are saying (2)

Newest first

Most points

Source	Message	Score	Date
	ROCmFix and InferBench – AMD Local-LLM Setup and Vulkan vs. Hip Benchmarking https://github.com/xanpavle/rocmfix/blob/main/PRIVACY.md#how... It should be opt in. It should always be opt in.	0	Sep 19, 2026
	ROCmFix and InferBench – AMD Local-LLM Setup and Vulkan vs. Hip Benchmarking	2	Sep 19, 2026

README

🔧 ROCmFix

Auto-detect your AMD GPU, fix ROCm overrides, and benchmark your AI backends in seconds.

One file. Zero installs. No pip. No venv. Just run it.

❓ What is this?

Running local AI (llama.cpp, Ollama, LM Studio, PyTorch) on AMD GPUs often leads to cryptic errors like HSA_STATUS_ERROR_INVALID_ISA or silent fallbacks to CPU. Usually, you just need to set the HSA_OVERRIDE_GFX_VERSION environment variable.

ROCmFix detects your exact AMD GPU, checks a live community database, automatically applies the correct shell environment override, and runs health diagnostics to ensure your setup is ready.

✨ Features

Auto-Apply & Undo: Automatically modifies CMD, PowerShell, Bash, Zsh, or Fish configs to apply overrides safely.
rocmfix doctor: Scans your hardware, Adrenalin drivers, HIP SDK, and Vulkan API to diagnose issues.
rocmfix bench: A 10-second backend race! Temporarily isolates Ollama or LM Studio, runs a generation test using Vulkan, then HIP, and tells you which is faster on your PC.
rocmfix install-hip: Missing the HIP SDK on Windows? This safely downloads and launches the official AMD installer for you.
Live Database & Self-Updater: Fetches new GPUs from the cloud daily, and updates itself via rocmfix update.
Export System Report: Generate a Markdown file of your setup for Reddit/Discord help posts.

🚀 Quick Start

Download (One Command)

Windows PowerShell:

Invoke-WebRequest -Uri "https://raw.githubusercontent.com/xanpavle/rocmfix/main/rocmfix.py" -OutFile "rocmfix.py"

Linux / macOS:

curl -O https://raw.githubusercontent.com/xanpavle/rocmfix/main/rocmfix.py

Run

python rocmfix.py

📋 Commands

Command	What it does
`rocmfix`	Detect GPU, suggest override, and auto-apply
`rocmfix doctor`	Full health check of drivers, HIP, and Vulkan
`rocmfix bench`	Benchmark Vulkan vs HIP/ROCm in LM Studio or Ollama
`rocmfix install-hip`	Download and install the AMD HIP SDK (Windows)
`rocmfix export`	Generate a Markdown system report for troubleshooting
`rocmfix update`	Update ROCmFix to the latest version from GitHub
`rocmfix sync`	Force-fetch the latest GPU database from the community server
`rocmfix test`	Smoke test current override to see if ROCm/HIP detects it
`rocmfix undo`	Safely revert the last auto-applied override
`rocmfix list`	Show all 17+ GPUs currently supported in the database
`rocmfix telemetry`	Manage your anonymous benchmark sharing settings
`rocmfix install`	Install `rocmfix` as a global command on your system

🎮 Supported GPUs

ROCmFix supports RDNA4 (RX 9070/9060 series), RDNA3 (RX 7000 series), RDNA2 (RX 6000 series), and Integrated Graphics.

New GPUs are synced automatically from our live database without needing to update the app!

Don't see your card? Run python rocmfix.py — it will generate a pre-filled GitHub issue link so you can contribute your GPU!

🔥 HIP vs Vulkan — Which should I use?

If you're running llama.cpp / LM Studio / Ollama on a consumer AMD card (especially on Windows):

	Vulkan	HIP / ROCm
Setup	✅ Zero config	⚠️ Needs HIP SDK + overrides
Windows	✅ Works out of the box	⚠️ HIP SDK still maturing
Linux	✅ Good	✅ Better
Consumer RDNA3 perf	✅ Often faster	⚠️ Sometimes slower
PyTorch support	❌ Limited	✅ Full support

TL;DR

Running LLMs on Windows? → Try Vulkan first. Run rocmfix bench to verify.
PyTorch / Stable Diffusion? → Run rocmfix install-hip, then apply the override ROCmFix gives you.

📊 Privacy & Telemetry

ROCmFix includes an opt-in telemetry system to help grow the community database. If enabled, ROCmFix securely sends your GPU model, driver version, OS, and benchmark results to a private collector. We never collect usernames, IPs, file paths, or AI prompts. Read the full details in PRIVACY.md.

📄 License

This project is licensed under the MIT License - see the LICENSE file for details.

If this saved you hours of Reddit digging, drop a ⭐
Made for the local AI community by xanpavle

Contributors

xanpavle

9 commits

Similar repos

sebastianmechno-sys/vllm-rocm-windows-rdna2

Native vLLM and ROCm 7.x runtime acceleration for AMD Radeon RX 6000 Series (RDNA2) on Windows 11.

Speedstu/CUDA-for-AMD-Windows

Run CUDA-targeted Windows applications on AMD GPUs with ZLUDA + ROCm/HIP.

266

lamikr/rocm_sdk_builder

445

charlie12345/vLLM_for_AMD

vLLM optimized for AMD

kyuz0/amd-strix-halo-toolboxes

1,943

scooter-lacroix/Stan-s-ML-Stack

A complete package that provides you with all the components needed to get started of dive deeper…

xynexus/hipfire

RDNA-native LLM inference engine in Rust.

ROCm/ROCm

AMD ROCm™ Software - GitHub Home

6,817

Languages

Python

100.0%

xanpavle/rocmfix

Auto-detect AMD GPUs on Windows/Linux and instantly get the correct HSA_OVERRIDE_GFX_VERSION for local AI (ROCm/HIP).

Python

9 commits

updated Sep 19, 2026

See the code

See what people are saying (2)

Newest first

Most points

Source	Message	Score	Date
	ROCmFix and InferBench – AMD Local-LLM Setup and Vulkan vs. Hip Benchmarking https://github.com/xanpavle/rocmfix/blob/main/PRIVACY.md#how... It should be opt in. It should always be opt in.	0	Sep 19, 2026
	ROCmFix and InferBench – AMD Local-LLM Setup and Vulkan vs. Hip Benchmarking	2	Sep 19, 2026

README

🔧 ROCmFix

Auto-detect your AMD GPU, fix ROCm overrides, and benchmark your AI backends in seconds.

One file. Zero installs. No pip. No venv. Just run it.

❓ What is this?

ROCmFix detects your exact AMD GPU, checks a live community database, automatically applies the correct shell environment override, and runs health diagnostics to ensure your setup is ready.

✨ Features

Auto-Apply & Undo: Automatically modifies CMD, PowerShell, Bash, Zsh, or Fish configs to apply overrides safely.
rocmfix doctor: Scans your hardware, Adrenalin drivers, HIP SDK, and Vulkan API to diagnose issues.
rocmfix bench: A 10-second backend race! Temporarily isolates Ollama or LM Studio, runs a generation test using Vulkan, then HIP, and tells you which is faster on your PC.
rocmfix install-hip: Missing the HIP SDK on Windows? This safely downloads and launches the official AMD installer for you.
Live Database & Self-Updater: Fetches new GPUs from the cloud daily, and updates itself via rocmfix update.
Export System Report: Generate a Markdown file of your setup for Reddit/Discord help posts.

🚀 Quick Start

Download (One Command)

Windows PowerShell:

Invoke-WebRequest -Uri "https://raw.githubusercontent.com/xanpavle/rocmfix/main/rocmfix.py" -OutFile "rocmfix.py"

Linux / macOS:

curl -O https://raw.githubusercontent.com/xanpavle/rocmfix/main/rocmfix.py

Run

python rocmfix.py

📋 Commands

Command	What it does
`rocmfix`	Detect GPU, suggest override, and auto-apply
`rocmfix doctor`	Full health check of drivers, HIP, and Vulkan
`rocmfix bench`	Benchmark Vulkan vs HIP/ROCm in LM Studio or Ollama
`rocmfix install-hip`	Download and install the AMD HIP SDK (Windows)
`rocmfix export`	Generate a Markdown system report for troubleshooting
`rocmfix update`	Update ROCmFix to the latest version from GitHub
`rocmfix sync`	Force-fetch the latest GPU database from the community server
`rocmfix test`	Smoke test current override to see if ROCm/HIP detects it
`rocmfix undo`	Safely revert the last auto-applied override
`rocmfix list`	Show all 17+ GPUs currently supported in the database
`rocmfix telemetry`	Manage your anonymous benchmark sharing settings
`rocmfix install`	Install `rocmfix` as a global command on your system

🎮 Supported GPUs

ROCmFix supports RDNA4 (RX 9070/9060 series), RDNA3 (RX 7000 series), RDNA2 (RX 6000 series), and Integrated Graphics.

New GPUs are synced automatically from our live database without needing to update the app!

Don't see your card? Run python rocmfix.py — it will generate a pre-filled GitHub issue link so you can contribute your GPU!

🔥 HIP vs Vulkan — Which should I use?

If you're running llama.cpp / LM Studio / Ollama on a consumer AMD card (especially on Windows):

	Vulkan	HIP / ROCm
Setup	✅ Zero config	⚠️ Needs HIP SDK + overrides
Windows	✅ Works out of the box	⚠️ HIP SDK still maturing
Linux	✅ Good	✅ Better
Consumer RDNA3 perf	✅ Often faster	⚠️ Sometimes slower
PyTorch support	❌ Limited	✅ Full support

TL;DR

Running LLMs on Windows? → Try Vulkan first. Run rocmfix bench to verify.
PyTorch / Stable Diffusion? → Run rocmfix install-hip, then apply the override ROCmFix gives you.

📊 Privacy & Telemetry

📄 License

This project is licensed under the MIT License - see the LICENSE file for details.

If this saved you hours of Reddit digging, drop a ⭐
Made for the local AI community by xanpavle

Similar repos

sebastianmechno-sys/vllm-rocm-windows-rdna2

Native vLLM and ROCm 7.x runtime acceleration for AMD Radeon RX 6000 Series (RDNA2) on Windows 11.

Speedstu/CUDA-for-AMD-Windows

Run CUDA-targeted Windows applications on AMD GPUs with ZLUDA + ROCm/HIP.

266

lamikr/rocm_sdk_builder

445

charlie12345/vLLM_for_AMD

vLLM optimized for AMD

kyuz0/amd-strix-halo-toolboxes

1,943

scooter-lacroix/Stan-s-ML-Stack

A complete package that provides you with all the components needed to get started of dive deeper…

xynexus/hipfire

RDNA-native LLM inference engine in Rust.

ROCm/ROCm

AMD ROCm™ Software - GitHub Home

6,817

Contributors

xanpavle

9 commits

Languages

Python

100.0%