Show HN: Navatala GPU – multi-back end GPU kernels and Python bindings

Category: library

Tags: gpu-compute, scientific-computing, cross-platform

Score: 6.5/10 (Innovation: 6, Technical: 8, Documentation: 7, Utility: 5)

Navatala GPU provides a portable GPU compute runtime and kernel corpus supporting multiple backends (CUDA, HIP, Vulkan, OpenCL, Metal), targeting scientific computing and CFD workloads. Its unified API and cross-platform kernel design are technically ambitious, but the project is in early alpha with uneven backend coverage and limited adoption.

Target audience: backend devs, data engineers, hpc scientists

Repository: https://github.com/navatala-systems/navatala_gpu · C++ · Apache-2.0

View on Hacker News