amd

H.O.M.E.T Una aplicación para optimizar las temperaturas y el rendimiento de CPUs.

Knowledge Integration & Response Agent v1.3

RAM-Backed MCP Memory Architecture for Consumer LLM Inference — 900K token context on 16GB VRAM

Tweak is Mwk Ambrose provided to help you increase your device's performance in a safe and free💯way. Follow us to get everything new.

Ryzentosh! Credits to @UniversoHackintosh by the helpful content.

Ryzentosh - Hackintosh based on TRX40-E and Threadripper 3960x

Bulk translate source code files from one language to another using Ollama LLM API.

Disables USB power saving to prevent crashes and BSODs on AMD systems.

Here you can find all my FPGA projects that I have worked on.

RedFFmpegatron is an intuitive Windows GUI application designed to streamline video processing tasks. It integrates FFmpeg with AMD’s AMF for high-per...

Open-source single-image to 3D reconstruction, optimized for consumer AMD GPUs

AMD/NVIDIA GPU cluster infrastructure — ~300 GPU deployment, ROCm, kernel tuning, multi-node benchmarking

🧭 Enhance navigation with VLN-YuanNav, a visual-language model using advanced memory and decision-making for effective exploration.

NVMe SSD M.2 M Key PCIe Card Carrier for Xilinx FPGA with FMC and FMC+

macOS on Gigabyte Z490M Gaming X - i7-10700K & Sapphire Pulse AMD Radeon RX 6600 XT

Ryzen Master Localization Repo

Benchmark script for llama.cpp & results for AMD RX 7900 XTX

Universal gaming optimisation suite for all Debian-based Linux distros — AMD, NVIDIA and Intel Arc

R.A.G-Race-Router [TAP — Trigonometric Assembly Processing Engine] — Self-optimizing CPU+iGPU+NPU inference via cosine similarity routing on AMD Ryzen...

A lightweight TUI monitor for AMD Ryzen AI NPUs

Sticky-block topology lottery scheduler for transformer fine-tuning.

Faster-Whisper on AMD GPUs via DirectML on Windows — drop-in GPU transcription, no ROCm required

A multi-agent research assistant for AMD hardware. Leverages Lemonade Omni Models to orchestrate parallel RAG, Web, Code, and Vision workers entirely...

High-sampling-rate, high-memory-density, multi-channel arbitrary waveform generator implemented on the RealDigital 4x2 RFSoC

Linux kernel patches enabling speaker output on Huawei MateBook (HVY-WXX9 / M1080). AMD Renoir + ES8316 codec.

Fine-tune LLMs on an AMD GPU on Windows via DirectML — no NVIDIA, no ROCm, no dual boot. Proven with a Mistral-7B LoRA fine-tune through Axolotl on a...

🍏 Comprehensive OpenCore Hackintosh installation guide — hardware compatibility matrix, EFI building, BIOS settings, troubleshooting, post-install op...

Real-time WGSL compute shaders on AMD iGPU → TouchDesigner Shared Memory In TOP or NDI. 4K @ 60+ fps with a live HTML control panel. 16 shaders includ...

Agentic calling Setup

App to handle one GPU to multiple models/element with queuing and log tracking all with a local hosted web dashboard

Linux driver + GUI configurator for the Thermaltake TH420 V2 Ultra EX ARGB round TFT LCD (480×480). Protocol fully reverse-engineered.

ComfyUI nodes for AMD AMF VideoSR/HQ video upscaling through FFmpeg sr_amf.

Evaluation-backed AMD ROCm port of HunyuanOCR-1.5 with reproducible OmniDocBench v1.6 benchmarks across llama.cpp, vLLM, and Transformers on gfx1100.

Reproducible AMD ROCm evaluation and full three-camera recording for BiGym, OpenPI JAX, MuJoCo and 3DGS.

Fixes random reboots/crashes on Ryzen Zen3 laptops (5900HX, etc.) from C6 idle-sleep wake failures — keeps every core out of deep sleep.

Rust-based AI model inference library for fast and efficient local model serving.

The newcomer's bible for the ASRock AMD BC-250 — a PS5-derived APU board turned into a budget Linux gaming & AI mini PC. 15 languages.

High-performance FP32 GEMV for AMD RDNA 3 (gfx1100), with reproducible performance and numerical analysis.

Tuning llama.cpp's Vulkan backend for AMD Vega 56/64 (gfx900/GCN5): 1-bit and MoE inference kernels, multi-GPU layer-split findings, and the dead ends

Rewind a training run to the instant its loss spiked, localize the poisoned optimizer state, repair just that slice, and prove it via bit-identical re...

AMD BC-250 CPU power management on Linux

AMD/HIP compatibility fixes and GPU-side autoregressive optimizations for Fish Audio S2 Pro on GGML-based C++ runtimes.

PAIR Universal BETA — universal fork of NVIDIA Personal AI Router, supports all GPUs (GTX/Tesla/AMD/Intel/Apple Silicon) + all Linux/macOS. Apache-2.0...

1bit engine: local LLM inference for AMD Ryzen AI (Strix Halo) inside Lemonade. HRX with our Loom kernels on the Radeon iGPU, plus the XDNA 2 NPU; Ope...

Source for the ROCm documentation website (rocm.docs.amd.com)

Early Attestation Considered Very Harmful (CVE-2026-92701, CVE-2026-92702, CVE-2026-33697, and more to come)

MiniMax H3 video and audio generation on AMD Strix Halo, powered by Loom and HRX.

Open Windows (WDDM) driver stack for AMD GPUs, developed first on the ASRock BC-250 (gfx1013). Not affiliated with AMD, ASRock or Microsoft.

Usability benchmark for local LLMs on AMD hardware (RDNA4 + Strix Halo) — agentic throughput, self-repair counts and context-depth decay, all parsed f...

AMD Radeon GPU-PV for real Linux Hyper-V VMs — D3D12/OpenGL through dxgkrnl with automatic WSL/UMD sync and GPU readback proof.