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 DeepSeek-R1, gemini-4-argon_high, gpt-6-astra_max and 7 more, by GPQA Diamond.
Create it free
Takes a free account
DeepSeek-R1
Aider Polyglot
57 %
Algotune
2 %
Arc Agi
16 %
Arc Agi 2
1 %
BALROG
35 %
Critpt
1 %
GPQA Diamond
69 %
Lech Mazur Writing
8 %
Live bench
72 %
Math Level 5
93 %
OTIS Mock AIME 2024 2025
53 %
SciCode
38 %
SimpleBench
31 %
WeirdML
36 %
Training Compute
3,500,000,000,000.00T FLOP
DeepSeek-R1 benchmark results.