News for hackers. Stuff that matters.
Re:Kimi K3.1 Pro and K3.1 Pro Highspeed by minimaxir on Friday October 23, 2026 @01:39PM (#50212657)
This is a price drop from Kimi K3 ($3/$15/$0.30) despite being a slightly larger model (3T vs. 2.8T params).

I'm glad I waited a day to post my analysis of LLM price cuts because it needs...an addition.

Re:Kimi K3.1 Pro and K3.1 Pro Highspeed by Philpax on Friday October 23, 2026 @01:44PM (#50212727)
Are we sure this is a new pretrain and not them rounding up 2.8T?
Re:Kimi K3.1 Pro and K3.1 Pro Highspeed by vikramkr on Friday October 23, 2026 @01:53PM (#50212850)
Literally from the linked announcement:

> Kimi K3.1 Pro is a post-trained version of Kimi K3, with substantially larger post-training compute

Re:Kimi K3.1 Pro and K3.1 Pro Highspeed by minimaxir on Friday October 23, 2026 @01:45PM (#50212757)
The Kimi K3 announcement (https://www.kimi.ai/blog/kimi-k3) used 2.8T and it would be weird to round that up for this announcement.
Re:Kimi K3.1 Pro and K3.1 Pro Highspeed by Philpax on Friday October 23, 2026 @02:06PM (#50213053)
Looks like this post's using "3T-class", which I think I've seen them use before.
Re:Kimi K3.1 Pro and K3.1 Pro Highspeed by rfoo on Friday October 23, 2026 @01:47PM (#50212775)
it's almost certainly rounding up 2.8T
Re:Kimi K3.1 Pro and K3.1 Pro Highspeed by tonyhart7 on Friday October 23, 2026 @02:32PM (#50213420)
lol, price mtok mean nothing if they substantially generate more reasoning token that would just add more cost to the task
Re:Kimi K3.1 Pro and K3.1 Pro Highspeed by minimaxir on Friday October 23, 2026 @02:33PM (#50213432)