"At 1M tokens, SubQ 1.1 Small requires 64.5x less compute than dense attention a... | Hacker News

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

projektfu 35 days ago | parent | context | favorite | on: SubQ 1.1 Small

"At 1M tokens, SubQ 1.1 Small requires 64.5x less compute than dense attention and runs 56x faster than FlashAttention-2."

6450% less compute? Is Trump working there?

Consider applying for YC's Fall 2026 batch! Applications are open till July 27.
Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact