rdna4

Star

Here are 56 public repositories matching this topic...

Language: All

Filter by language

All 56 Python 23 C++ 9 Shell 8 Cuda 4 Rust 3 PowerShell 2 Zig 2 HTML 1 JavaScript 1

Sort: Most stars

Sort options

Most stars Fewest stars Most forks Fewest forks Recently updated Least recently updated

zinc

zolotukhin / zinc

Star 513

Zig INferenCe Engine — Local LLM inference on AMD GPUs and Apple Silicon

amd gpu zig pytorch transformer openai gpt amdgpu rdna3 rdna4 qwen3

Updated Sep 8, 2026
Zig

Windows-only version of ComfyUI which uses AMD's official ROCm and PyTorch libraries to get better performance with AMD GPUs. [auto-installation and popular performance enhancing packages like triton * sage-attention * flash-attention * bitsandbytes included ]

windows triton rdna rocm miopen bitsandbytes flash-attention rdna3 rdna2 rdna4 sage-attention rdna1

Updated Sep 7, 2026
Python

0ptiscaler4linux

ind4skylivey / 0ptiscaler4linux

Star 80

The intelligent OptiScaler installer Linux gamers needed. Automates FSR4, XeSS & DLSS configuration with GPU-optimized profiles for RDNA3/4, Arc & RTX cards.

vulkan proton linux-tools shell-scripting upscaling mesa gaming-performance gpu-optimization dlss linux-gaming xess steam-deck rdna3 frame-generation rdna4 optiscaler fsr4 amd-fsr

Updated Aug 4, 2026
Shell

GPUOpen-Tools / isa_spec_manager

Star 57

Utilities for accessing AMD's Machine-Readable GPU ISA Specifications.

amd gpu specification isa rdna cdna mi300 rdna3 rdna2 cdna3 cdna2 rdna4

Updated Aug 7, 2026
C++

Dyluhn / R9V

Star 29

Shape-specialized ROCm inference for Qwen3.8 Flash Next on RDNA4

rocm vllm qwen gguf speculative-decoding rdna4

Updated Sep 4, 2026
Rust

miklebel / adrenalift

Star 23

Windows tool to bypass AMD driver clock gating and unlock real GPU boost clocks (RDNA4)

windows amd gpu overclock amdgpu powerplay overclocking rdna4

Updated Jun 7, 2026

maeddesg / vulkanforge

Star 21

LLM inference engine for AMD RDNA4 — Rust + Vulkan compute shaders, gguf & native FP8.

rust machine-learning amd vulkan inference mesa llm fp8 gguf rdna4 gfx1201 gemma4

Updated Jun 20, 2026
Rust

buptanswer / mineru

Star 13

让 MinerU 在 AMD 显卡上以 vLLM + hybrid-auto-engine 满血运行! 本项目是对官方文档中 AMD/ROCm 生态支持的有效补充。针对 MinerU 3.x 提供了一套完整的 ROCm 7.x + PyTorch 2.11.0 + vLLM 源码编译与部署方案,解析质量与速度对标 NVIDIA 旗舰显卡。

python pytorch data-extraction rocm e-book amd-gpu rag wsl2 llm document-parsing vllm mineru rdna4

Updated May 27, 2026
Python

KaiFelixBennett / gemma4-turboquant-rdna4

Star 12

Run Gemma-4-31B at full 256K context on a 1,400ドル AMD RDNA4 GPU (gfx1201): TurboQuant KV cache + HIP-graph-safe Flash-Attention for llama.cpp, fully measured on real hardware.

amd hip quantization gemma rocm kv-cache long-context llama-cpp local-llm llm-inference flash-attention rdna4 gfx1201 turboquant

Updated Jul 23, 2026
Python

daimonionnn / amd-r9700-vllm-and-tuning-toolkit

Star 12

AMD Radeon 9700 AI PRO RDNA multi GPU Tools for llm/vLLM inference, tuning, overclocking, undervolting and LLM models benchmarking

linux benchmark ubuntu amd rocm radeon vllm tensorparallel rdna4 amd-radeon gfx1201

Updated Aug 15, 2026
Shell

Capicua25x / vllm-rocm-rdna4

Star 11

vLLM for AMD RDNA4 (gfx1201): Radeon AI PRO R9700 & RX 9070 XT — native MXFP4 linear kernel + spec-decode verify fix, carried as rebased branches (fork-carry model)

amd rocm vllm llm-inference rdna4 mxfp4 gfx1201 r9700

Updated Sep 3, 2026
Python

tlee933 / llama.cpp-rdna4-gfx1201

Star 11

llama.cpp with native AMD RDNA4 (gfx1201) ROCm 7.11 support - 98.97 tok/s AI inference, competitive with RTX 4070 Ti, 32GB VRAM

machine-learning amd gpu hip rocm ai-inference llm llama-cpp rdna4 gfx1201

Updated Jan 3, 2026
C++

MrLordCat / llama.cpp-rdna-lab

Star 6

Hardware-focused llama.cpp fork for Windows and dual AMD RDNA4 RX 9070 XT GPUs: ROCm/HIP + Vulkan backends, PyQt6 GUI (RDNA LLM Studio), MTP speculative decoding, FP8 attention, long-context Qwen3.8 benchmarks

windows benchmark amd vulkan rpc gpu-acceleration hip mtp rocm fp8 llama-cpp ggml llm-inference qwen agentic-ai rdna4 qwen3 gfx1201 rx-9070-xt

Updated Sep 7, 2026
C++

RadeonForge

KaiFelixBennett / RadeonForge

Star 6

Fine-tune your own LLM on an AMD Radeon GPU — the easy, tested way. QLoRA via ROCm on Windows/WSL2 & Linux, a worked Gemma-4 example, a reusable live training dashboard, and a smoke test that proves the loss falls.

amd pytorch lora gemma rocm radeon fine-tuning peft wsl2 llm llama-cpp qlora rdna4 gfx1201

Updated Jun 20, 2026
HTML

Rolaand-Jayz / RE-of-FSR-4.1.0-Upscaling

Star 5

Static structural analysis of AMD FSR 4.1.0 neural upscaler: DXIL entrypoint inventory, weight extraction, blob layout, strict verification, and rebuild section comparison for interoperability research.

open-source machine-learning neural-network amd gpu directx reverse-engineering upscaling fsr rdna4 fsr4

Updated Jul 2, 2026
Python

hazyumps / vllm-gfx1201-qwen3.6-256k

Star 4

Qwen3.6-27B-FP8 at 256k context on dual AMD RDNA4 (gfx1201) — working vLLM recipe, audited measurement harness, and the full debugging record

triton rocm amd-gpu long-context fp8 vllm llm-inference qwen rdna4 gfx1201

Updated Aug 7, 2026
Python

somestupidgirl / RDNA4FB

Sponsor

Star 4

RDNA4 GPU driver stuff for macOS

macos osx hackintosh kext amdgpu radeon ryzentosh rdna4

Updated Aug 21, 2026
C++

AlanHuang99 / qwen3.6-mtp-stack

Star 4

Reproducible Docker + LiteLLM stack for running Qwen 3.6 27B with MTP on AMD RDNA4 (gfx1201). Verified on Radeon AI PRO R9700 32 GB. Built from llama.cpp PR #22673.

amd vulkan homelab llama-cpp qwen speculative-decoding litellm rdna4