#
rdna3
Here are 52 public repositories matching this topic...
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 ]
-
Updated
Sep 7, 2026 - Python
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
TRELLIS (Microsoft's Image-to-3D generator) running on AMD GPUs with ROCm. Includes Gaussian splatting, mesh extraction, and GLB export. Tested on RX 7800 XT.
trellis hip rocm 3d-generation amd-gpu image-to-3d amd-gpus gaussian-splatting rdna3 mesh-extraction nvdiffrast
-
Updated
May 9, 2026 - Jupyter Notebook
A turnkey, fully-local AI workstation engineered for the AMD Ryzen AI Max+ 395. LLM inference, voice, document parsing, browser automation, agents — all on-device.
docker-compose amd vulkan self-hosted browser-automation rocm unified-memory ai-agent llm llama-cpp local-ai qwen litellm rdna3 ai-workstation strix-halo gfx1151 ryzen-ai-max radeon-8060s ryzen-ai-max-plus-395
-
Updated
Jul 30, 2026 - Python
A honest port of vLLM-ROCm for windows.
-
Updated
Aug 23, 2026 - Python
Reproducible RDNA reference for Meta Muse-Glimmer-30B — adapting MI-series ROCm recipes to Ryzen AI and Radeon with vLLM, llama.cpp, DFlash and auditable benchmarks.
benchmarking amd rdna reproducibility rocm radeon llama-cpp vllm gguf speculative-decoding rdna3 ryzen-ai strix-halo gfx1151 muse-glimmer
-
Updated
Aug 17, 2026 - Python
Evaluation-backed AMD ROCm port of HunyuanOCR-1.5 with reproducible OmniDocBench v1.6 benchmarks across llama.cpp, vLLM, and Transformers on gfx1100.
-
Updated
Sep 7, 2026 - Python
Unlock fast, local LLM inference on AMD-powered mini PCs delivering 65-87 t/s for large models without cloud or subscription costs
amd optimization inference rocm mini-pc asus-rog linux-gaming unified-memory beelink cachyos llm llama-cpp local-llm ollama gguf rdna3 strix-halo gfx1151
-
Updated
Sep 6, 2026 - Shell
Multi-GPU vLLM on consumer Radeon (RX 7900 XT/XTX, gfx1100): root cause and fix for the RCCL "operation cannot be performed in the present state" crash, plus 292 benchmark measurements across five model architectures
benchmark hip rccl multi-gpu vfio rocm radeon amd-gpu gpu-passthrough tensor-parallelism vllm llm-inference gfx1100 rdna3 pcie-atomics
-
Updated
Sep 7, 2026 - Python
High-performance FP32 GEMV for AMD RDNA 3 (gfx1100), with reproducible performance and numerical analysis.
-
Updated
Aug 31, 2026 - C++
PyTorch built from source for AMD RDNA 3.5 (gfx1150) — Radeon 890M/880M GPU acceleration
-
Updated
Mar 30, 2026 - Shell
Working ROCm 7.2.x + PyTorch environment for RDNA3. Built after fighting pipeline rats and wondering, “Lisa Su, girl… what are they doing down there.”
machine-learning ai deep-learning amd transformers pytorch gpu-computing lora rocm radeon fine-tuning amd-gpu huggingface local-inference pytorch-rocm local-llm qlora rdna3 gfx1101
-
Updated
Jun 19, 2026 - Python
⚡ Zero-dependency Homebrew tap for high-performance LLMs (CachyLLama, ROCmFPX, llama-ai). Optimized for AMD ROCm 7 (Strix Halo, RDNA3/4), Apple Silicon Metal, & Vulkan RADV.
homebrew metal inference homebrew-tap quantization rocm amd-gpu apple-silicon ai-inference llm llama-cpp local-ai rdna3 rdna4 strix-halo strix-point rocmfpx rocm-gpu cachy-llama vulkan-radv
-
Updated
Sep 7, 2026 - Python
Own inference stack on AMD gfx1100: Mojo GPU kernels, self-describing GGUF engine
benchmark mojo amd gpu hip gemm rocm radeon inference-engine gpu-kernels matmul apache-2 llama-cpp llm-inference gfx1100 gguf rdna3 kernel-optimization wmma hipblaslt
-
Updated
Sep 6, 2026 - Mojo
Docker infrastructure for AMD Strix Halo (RDNA 3.5 / gfx1151): PyTorch + ROCm base container and a separate Ollama LLM service. Two folders, two Compose files, one Strix Halo box.
docker ubuntu docker-compose amd gpu self-hosted pytorch rocm llm ollama rdna3 ryzen-ai strix-halo gfx1151
-
Updated
Apr 26, 2026 - Shell
Multi-GPU tensor/context parallel diffusion on AMD ROCm — with the patch that makes it actually work.
-
Updated
Apr 19, 2026 - Python
Add this topic to your repo
To associate your repository with the rdna3 topic, visit your repo's landing page and select "manage topics."