Best score on CritPt benchmark (~FrontierMath for physics) by end of 2026
7
Ṁ250Ṁ1.5kDec 31
88.8 %
expected1H
6H
1D
1W
1M
ALL
0.6%
Below 20
0.6%
20 - 29
0.8%
30 - 39
0.9%
40 - 49
1.2%
50 - 59
1.2%
60 - 69
1.3%
70 - 79
33%
80 - 89
61%
90 - 100
https://critpt.com/
Best CritPt score at end of 2026.
Update 2026-07-10 (PST) (AI summary of creator comment): The creator will use https://artificialanalysis.ai/evaluations/critpt as the source for scores (instead of critpt.com) to ensure all models are included.
Update 2026-10-22 (PST) (AI summary of creator comment): The market will resolve using the corrected CritPt v2 evaluation at https://artificialanalysis.ai/evaluations/critpt-v2.
This question is managed and resolved by Manifold.
Get 
1,000 to start trading!
1,000Sort by:
I'll use the corrected version of critpt here for resolution: https://artificialanalysis.ai/evaluations/critpt-v2
What if the website is not updated (it is currenlty missing GPT 5.5 Pro, GPT 5.6, and Fable)? I suggest https://artificialanalysis.ai/evaluations/critpt which has all of them
People are also trading
Related questions
What will the top score on Epoch AI's FrontierMath: Erdős (68 open Erdős problems) be on November 7, 2026?
Highest score on Epoch AI's FrontierMath Erdős benchmark (68 open Erdős problems) by November 30, 2026?
Will Al achieve 95% or higher score on the FrontierMath benchmark before 2030?
99% chance
What will be true of the SOTA AI on the FrontierMath benchmark, before 2027?
Will a frontier model score above 90% on the APEX-SWE benchmark before 2028?
71% chance
In what year will Al achieve 95% or higher score on the FrontierMath benchmark?
2027
Benchmark Gap #8: Once a single AI gets >= 80% on FrontierMath Tier 4, how long until an AI publishes a math paper?
13