PFlash: 10x prefill speedup over llama.cpp at 128K on a RTX 3090
1–3 of 3 posts
Re: PFlash: 10x prefill speedup over llama.cpp at 128K on a RTX 3090
#2[dead]
Re: PFlash: 10x prefill speedup over llama.cpp at 128K on a RTX 3090
#3is this repo a legit one, has anyone audited it?