Users are excited by rapid progress in open-weight models that nearly close the performance gap with closed systems while respecting open principles.
Based on 6 visible X reactions from 9 accounts; directional sample.
Ask a question below.
Published answers will appear here.
The progress is driven by Kimi Delta Attention and MoE.
@jbhuang0604 finally a model that actually respects open-weight principles while pushing performance. attention residual is a neat trick—less compute, better memory.
@jbhuang0604 Thanks for sharing and I clicked on your profile and realized you work at UMD. Go terps! Definitely will subscribe to your channel
@jbhuang0604 I always relied on your videos when things gets confusing thanks <3
@jbhuang0604 actually good videos!
LLM Infra folks are having too much fun these days
Users are excited by rapid progress in open-weight models that nearly close the performance gap with closed systems while respecting open principles.
Based on 6 visible X reactions from 9 accounts; directional sample.
Ask a question below.
Published answers will appear here.