Benchmark Almanac

Benchmarks Arenas Leaders Models Labs Model cards New Sources Submit

/

Labs

Aorizon

1 models 0 model cards read

Every Aorizon model

HappyHorse 1.0

Benchmarks released each month since 2019, colored by area.

Benchmark Almanac

AI benchmarks, who leads them, and which ones labs still report.

2,721 benchmarks, 984 models and 103 labs. Data as of 2 Nov 2026.

Submit a benchmark

Explore

Benchmarks
Arenas
Leaders
Models
Labs
Model cards
New this month

Areas

Knowledge and language
Math and science
Coding
Agents and tools
Professional work
Vision, audio and robotics
Safety and security

Method

What gets cataloged
Where the scores come from
In use, fading, dropped
Versions
Who held number one
Finding new benchmarks
Sources

Scores from Epoch AI and Arena are under CC BY 4.0, and scores from other leaderboards are under each board’s own terms.