GitHub · C++
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.
- Stars
- 126
- Forks
- 16
- Stars gained in 24 hours
- +8
- Forks gained in 24 hours
- 0
Topics: cuda, gguf, llama-cpp, p100, pascal, quantization, speculative-decoding, v100, vllm, volta
Latest data point: Nov 22, 2026 · 126 stars · 16 forks