oivhtsm (@hardtosleepman) profile picture

oivhtsm

@hardtosleepman
 codex
Joined October 2026
  • Tweets10
  • Following2
  • Followers153
10Media
Photo by oivhtsm (@hardtosleepman)Photo by oivhtsm (@hardtosleepman)Photo by oivhtsm (@hardtosleepman)Photo by oivhtsm (@hardtosleepman)
It’s very hard to sleep man
METR (@METR_Evals) profile pictureMETR
@METR_EvalsNov 5
Replying to @METR_Evals
1. Benchmarks provide one piece of evidence for acceleration. Since Oct 2025, AECI scores for Anthropic’s models have grown at 1.9x [95%: 1.5–2.8] the earlier rate. Analysis choices affect this estimate, and it remains unclear how benchmark scores map to real-world capabilities.
Image from METR (@METR_Evals)
4
41
6,628
It's very hard to sleep man
1
2
28
17,150
It’s very hard to sleep man
2
2
1
42
10,595
It’s very hard to sleep man
2
6
19
2,915
It’s very hard to sleep man.
2
3
53
6,976
It's very hard to sleep man
Dan Roberts (@danintheory) profile pictureDan Roberts
@danintheoryOct 26
We are announcing three new discoveries in mathematics, computer science, and physics about harmonic analysis, complexity theory, and quantum field theory. openai.com/index/building-sc…
Working with academic communities to build scientific understanding

Working with academic communities to build scientific understanding

OpenAI shares model results on the Kakeya conjecture, Exponential Time Hypothesis, and quantum chromodynamics—and how researchers can verify and build on them.

openai.com
15
203
50,109
It's very hard to sleep man
1
5
1,668
It's very hard to sleep man
1
4
1,392
It's very hard to sleep man
1
6
1,244
Load more