The set-and-forget LLM engine for Pascal and Volta: PXQ codec + kernels, auto-tuned per card. Ready-to-run PXQ models in MODELS.md; benchmarks vs llama.cpp in the README.
↑ +24 today
Discover the best P100 open source repositories on GitHub. Explore trending tools, live star counts, source code, and developer projects updated daily.