Beledarian/wgpu-llm

A from-scratch LLM inference engine that uses wgpu (the cross-platform WebGPU implementation) to dispatch WGSL compute shaders for every math operation a Transformer needs. No CUDA. No Python. No massive framework dependencies. Just Rust, raw shaders, and your GPU.

2 best points · 2 comments · 1 Hacker News post · 13 GitHub stars · View on GitHub

Hacker News discussions

• I built a pure WGSL LLM engine to run Llama on my Snapdragon laptop GPU (2 points, 2 comments, by Beledarian, 2026-04-11)