back
user profile
teleforce
22,692karma·6,478submissions·October 24, 2019
recent activity (6,478 total)
2 pts
comment
>This repository provides a patch for SGLang and vLLM that enables IndexCache inference acceleration for models using DeepSeek Sparse Attention (DSA), including DeepSeek-V3.2 and GLM-5. Paper here …
2 pts
4 pts
comment
> Call it AI-first, AI-proficient, whatever you like Can we just call it AI assistant and since it is really what it is. Just call a spade a spade, call it a day. Nvidia boss Jensen Huang refer to …
comment
I think the replies were pre-mature since when I commented on HN the comments total still has not reached 1000, but he already linked the replies already in the blog. Perhaps should have waited the co…