Leaderboards
Greatness isn't accidental. How we measure it shouldn't be either. If we want AGI that builds billion-dollar enterprises and globe-spanning infrastructure, we can't evaluate it with clickbait and synthetic slop. We need benchmarks that test for intelligence and sophistication.
This is our definitive ranking of models, measured by their capacity for rigorous reasoning and real-world mastery. Discover which labs are leading the frontier.
This is our definitive ranking of models, measured by their capacity for rigorous reasoning and real-world mastery. Discover which labs are leading the frontier.
Creative, Business, and Everyday writing
Hemingway-bench
Stop rewarding slop.
We take real-world writing tasks and put them in front of master wordsmiths.
Our goal: to push AI writing from two-second vibes to genuine nuance and impact.
Rank
Model
elo score (95% ci)
Gemini 3.1 Pro
1093
(
1072
-
1114
)
Google
Gemini 3 Flash
1088
(
1070
-
1106
)
Google
Gemini 3 Pro
1082
(
1060
-
1105
)
Google
Claude Opus 4.6
1064
(
1043
-
1086
)
Anthropic
Claude Opus 4.5
1047
(
1028
-
1066
)
Anthropic
Claude Sonnet 4.6
1032
(
1012
-
1053
)
Anthropic
GPT-5.2 Chat
1029
(
1010
-
1048
)
OpenAI
Kimi K2.5
1029
(
1010
-
1049
)
Moonshot AI
GPT-5.4
1022
(
999
-
1044
)
OpenAI
Qwen3.5 Plus
1008
(
988
-
1028
)
Alibaba Cloud
GPT-5.2
990
(
967
-
1013
)
OpenAI
Qwen3 Max
983
(
958
-
1008
)
Alibaba Cloud
View full leaderboard
Enterprise Agents in Realistic RL Environments
EnterpriseBench: CoreCraft
Stop testing models in tiny, self-contained environments.
We built CoreCraft, a large-scale startup world, and deployed AI agents to solve real tasks.
Our goal: to move agents beyond the cleanliness of the lab and into the chaos of enterprise reality.
Rank
Model
Score
GPT-5.2 (xHigh reasoning)
42.6
%
OpenAI
GPT-5.4 (xHigh reasoning)
36.4
%
OpenAI
Claude Opus 4.6 (Adaptive Thinking + Max Reasoning Effort)
30.8
%
Anthropic
View full leaderboard
Mathematics at the frontier
Riemann-bench
We evaluate AI models on advanced mathematical problems requiring deep reasoning and novel synthesis. Our benchmark features problems from cutting-edge mathematics, sourced from leading mathematicians – Ivy League professors, PhD IMO medalists, graduate students at the top of their field – in the course of their research.
Rank
Model
Score
Gemini 3.1 Pro
6
%
Google
Claude Opus 4.6
6
%
Anthropic
Gemini 3 Pro
4
%
Google
Kimi K2.5
4
%
Moonshot AI
DeepSeek v3.2
3
%
Claude Opus 4.5
2
%
Google
GPT-5.2
2
%
OpenAI
View full leaderboard