1.
2.
3.
5.
6.
7.
Real-time LLM Inference on Standard GPUs: 3k tokens/s per request
(news.ycombinator.com)
8.
9.
11.
12.
13.
14.
15.
Why isn't AMD's MI300X competitive?
(news.ycombinator.com)
16.
17.
18.
Is a $30,000 GPU Good at Password Cracking?
(bleepingcomputer.com)
19.
20.
Scaling Karpathy's Autoresearch: What Happens When the Agent Gets a GPU Cluster
(news.ycombinator.com)
21.
Today's top topics:
anthropic
apple
openai
google
ai safety
iphone duo
ios 27
iphone 18 pro
dario amodei
android