1.
2.
A New Trick Reveals AI Models’ Inner Thoughts
(wired.com)
3.
DDoS attacks over 1 Tbps surged fivefold in the second quarter
(bleepingcomputer.com)
4.
Itadakimasu: A word you say to the food, not the cook
(news.ycombinator.com)
5.
Itadakimasu: A Word You Say to the Food, Not the Cook
(news.ycombinator.com)
6.
Paper Goods Company Is Investigating an Alternative to Trees
(slashdot.org)
7.
9.
11.
Smaller, faster, safer: running Kimi and GLM at scale
(news.ycombinator.com)
12.
AirLLM 70B inference with single 4GB GPU
(news.ycombinator.com)
13.
What DMARC Protects You From, and What It Does Not
(news.ycombinator.com)
14.
Running Kimi K3 on MI355X at Better Performance per Dollar Than B300
(news.ycombinator.com)
15.
16.
Run Kimi K3 using 29 GB of RAM at 0.50 tok/s
(news.ycombinator.com)
17.
Moonshot’s Kimi uses 20k Nvidia chip cluster from Alibaba
(news.ycombinator.com)
20.
21.
Kimi K3-256k
(news.ycombinator.com)
22.
23.
Running Kimi K3 on a M1 Max
(news.ycombinator.com)
24.
Running Kimi K3 on a M1 Mac
(news.ycombinator.com)
25.
What to know about Moonshot AI and its new open-weight model Kimi K3
(feeds.feedburner.com)
26.
What to know about Moonshot AI and its new open weight model Kimi K3
(feeds.feedburner.com)
27.
The union drive at the Wikimedia Foundation is expanding
(theverge.com)
28.
A walk through of the DeltaNet family of linear attention variants
(news.ycombinator.com)
29.
You Could Have Come Up with Kimi Delta Attention
(news.ycombinator.com)
30.
Kimi K3 Architecture Overview and Notes
(news.ycombinator.com)