NHacker Next
  • new
  • past
  • show
  • ask
  • show
  • jobs
  • submit
▲LLM inference, visually explained: why a 13B model reads 26 GB per token (youtube.com)
4 points by ryansakhala 9 hours ago | 0 comments
Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact
Rendered at 08:10:09 GMT+0000 (Coordinated Universal Time) with Cloudflare Workers.