Ai Wire
CUDA+ROCm dual-back ends LLM inference for mixed-vendor GPUs (up to 48% faster)
Hacker News •
renwujun •
19h ago
Article URL: https://github.com/mw00/project-maya/pull/879
Comments URL: https://news.ycombinator.com/item?id=50435181
Points: 2
# Comments: 1
