ROCm GPU Computing
ROCm is an open software platform for programming and running workloads on graphics processing units. It provides tools, runtime components, and libraries that let applications use parallel processing for tasks such as scientific computing, data analysis, and machine learning. By moving suitable computations from general-purpose processors to GPUs, this software stack can help reduce processing time and support larger workloads, while giving developers ways to manage hardware access and integrate acceleration into existing code.
2 repositories · updated August 26, 2026
ds4: A Fast Local Inference Engine for DeepSeek V4 Flash and PRO on Metal, CUDA, ROCm
ds4 is a highly optimized, native inference engine designed for DeepSeek V4 Flash and PRO models. It provides efficient local inference across various hardware platforms, including Apple Silicon (Metal), NVIDIA GPUs (CUDA), and AMD ROCm. This project focuses on delivering high performance for large language models on consumer-grade machines.
AMD Skills: Empowering AI Agents with AMD's Optimized Software Stack
AMD Skills is the official catalog of AI agent skills from AMD, designed to empower AI agents with optimized software for AMD hardware. This repository provides knowledge, scripts, and conventions for working with AMD's stack, enabling seamless integration with major coding agents like Cursor, Claude Code, OpenAI Codex, and Gemini CLI.