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

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 ]

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

Installation script for an AI applications using ROCm on Linux.

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

Strix Halo local LLM guide: 63-97 t/s direct MoE on Ryzen AI MAX+ 395 / 128GB unified memory. Setup, model choices, benchmarks, and raw evidence.

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.

Multi-GPU tensor/context parallel diffusion on AMD ROCm — with the patch that makes it actually work.

FLUX.1-dev on AMD Radeon consumer GPUs — fast, low-VRAM, and shippable. Backport patches + benchmarks for torchao + diffusers group_offload on ROCm.

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.

MIOpen research for AMD RDNA 3.5 (gfx1150) — whitelist patch and three-bug analysis for Strix Point APU training

High-performance exact vector search on NVIDIA & AMD GPUs. 9354 QPS (RTX 5090D), 6275 QPS (RX 7900 XTX). Open-source, deterministic, and recall=1.0 — enabling Trustworthy AI at the edge.

Local LLM benchmarks on AMD Strix Halo — 26+ models tested across RADV, AMDVLK, and ROCm with llama.cpp

Local speech-to-text with AMD ROCm GPU acceleration

Visualize and monitor ROCm-SMI data locally or remotely via REST API — a more convenient alternative to rocm-smi.

Running LLMs locally on AMD GPUs, by people who actually do it. Cookbook + benchmark database for ROCm/HIP inference on Radeon, Instinct, and Strix Halo.

Enabling PyTorch on XLA Devices (e.g. Google TPU)

Hot-swappable multi-model LLM inference on a single AMD RX 7900 XTX (ROCm). 20+ model variants — instruct, thinking, vision, agent — orchestrated by llama-swap with VRAM eviction, MTP speculative decoding, and GPU power management.

Windows 11 gaming optimization for AMD Ryzen + RDNA 3. De-bloat, performance tweaks, app installation, AMD GPU guidance, game profiles. Interactive menu with dry-run, undo, and health checks.

Fixes and patches for NVlabs/nvdiffrast to support AMD ROCm 7.1 and Wave64 architectures (gfx1100, gfx1201).