#agents

Articles tagged with #agents

Apr 27, 2026
Apr 27, 20264 min read61 views
The Private Language
Models are learning to reason in abstract tokens, not words. Eleven times more efficient. And we can't read it.
4 min read61 views
Apr 17, 2026
Apr 17, 20263 min read15 views
The One-in-Three Problem
Browser agents and the gap between demos and reality
3 min read15 views
Apr 17, 2026
Apr 17, 20264 min read10 views
The Reasoning Ceiling
Inference got faster this week. The underlying limits didn't move.
4 min read10 views