poisonxa16/pxa

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.

语言：C++ · GitHub 星标：106 · Fork：15

话题：cuda · gguf · llama-cpp · models · p100 · pascal · pxq · quantization

在 GitHub 上查看原仓库