Some users praise the GLM-5.2 SSD streaming engineering on M5 Max as feeling like magic, while others sarcastically dismiss the speed claims or criticize researchers for correcting errors amid platform insults.
Based on 5 visible X reactions from 57 accounts; directional sample.
Ask a question below.
Published answers will appear here.
@francoisfleuret One of those cases where engineering feels like magic
@antirez @francoisfleuret you are writing history.
@antirez @francoisfleuret Bro that's sickening 😭😭
@francoisfleuret 在一个所有人都在造谣,骂人的地方,你居然在尝试修正你自己的错误。你不适合推特。
@francoisfleuret 10 seconds per token… “Look! It works!
I do not get it: - GLM-5.2 is 750B parameters, 40B active= 20Gb, - A top SSD can do *in theory* 15Gb/s. how can you do better than <1 token/s ?
Google response seems reasonable.
This kind of posts for instance, I wish I could summarize the responses I got and fix the mistake I wrote.
Some users praise the GLM-5.2 SSD streaming engineering on M5 Max as feeling like magic, while others sarcastically dismiss the speed claims or criticize researchers for correcting errors amid platform insults.
Based on 5 visible X reactions from 57 accounts; directional sample.
Ask a question below.
Published answers will appear here.