hip

HIP: C++ Heterogeneous-Compute Interface for Portability

RamaLama is an open-source developer tool that simplifies the local serving of AI models from any source and facilitates their use for inference in production, all through the familiar language of containers.

Stable Diffusion web UI

Vulkan/CUDA/HIP/OpenCL/Level Zero/Metal Fast Fourier Transform library

stdgpu: Efficient STL-like Data Structures on the GPU

Local inference that behaves like a frontier API: Qwen3.8-Flash-Next on AMD Strix Halo plus small NPU models, over the OpenAI and Anthropic APIs on one port.

RDNA-native LLM inference engine in Rust.

Run Time for AIE and FPGA based platforms

Numerical linear algebra software package

Peak compute, memory bandwidth and latency of CPUs, GPUs and NPUs: the same tests through every API that reaches a device. Command line and app for Windows, macOS, Linux and Android

🚀 Cenit IO - 100% open source integration Platform (iPaaS)

A GPU benchmark tool for evaluating GPUs and CPUs on mixed operational intensity kernels (CUDA, OpenCL, HIP, SYCL, OpenMP)

Portable and vendor neutral framework for parallel programming on heterogeneous platforms.

Abstraction Library for Parallel Kernel Acceleration 🦙

[DEPRECATED] Moved to ROCm/rocm-libraries repo

chipStar is a tool for compiling and running HIP/CUDA on SPIR-V via OpenCL or Level Zero APIs.

A curated learning repository focused on High-Performance Computing (HPC) — covering fundamentals to advanced topics in CUDA, MPI, C++, and Python-C++ interoperability.

Unbiased & physically-based GPU HIPRT (C++/HIP) interactive wavefront path tracing renderer

NVIDIA's official DLSS 3/4/4.5 on AMD RDNA3/4 GPUs on Windows and Linux.

Run CUDA-targeted Windows applications on AMD GPUs with ZLUDA + ROCm/HIP.