Hacker Newsnew | past | comments | ask | show | jobs | submitlogin
Speculative pre-positioning: off-path decode for stateful inference sessions (arxiv.org)
1 point by logotype 85 days ago | hide | past | favorite | 1 comment


With native support for sessions in an inference engine we can make use of idle GPUs... doing work!




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

Search: