This website uses cookies for authentication, support chat, and analytics. Only authentication cookies are turned on by default.
Animate this data
Bar chart race of Thinking Machines, Google DeepMind, OpenAI and 7 more, by GPQA Diamond.
Create it free
Takes a free account
Thinking Machines
Ale bench
946
Apex Agents
49 %
Arc Agi
84 %
Arc Agi 2
40 %
Chess Puzzles
21 %
Critpt
8 %
Dtbench
87 %
FrontierCode
14 %
FrontierMath Tier 4 V2
17 %
FrontierMath Tiers 1 3 V2
46 %
FrontierSWE
4 %
GPQA Diamond
89 %
Lmca
38 %
Mystery Game Puzzles
6 %
OTIS Mock AIME 2024 2025
90 %
Proofbench
6 %
SciCode
50 %
SimpleBench
50 %
SimpleQA Verified
40 %
WebDev Arena
1,410
WeirdML
32 %
WeirdML V3
4 %
Training Compute
11,000,000,000,000.00T FLOP
Thinking Machines benchmark results over time.