GGML Flash Attention support merged into llama.cpp | Hacker News

Hacker Newsnew | past | comments | ask | show | jobs | submit

		GGML Flash Attention support merged into llama.cpp (github.com/ggerganov)
		3 points by smcleod on April 30, 2024 \| hide \| past \| favorite \| 1 comment

Kelteseth on April 30, 2024 [–]

Looks like this is based on https://arxiv.org/abs/2205.14135

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact