o4-mini
#83 of 274 scored.
OpenAIApr 16, 2025Closed
Capabilities Index
Range 143.2–147.4
145.6
Context
Takes image, text, file
200K
Input
Per 1M tokens
$1.10
Output
Per 1M tokens
$4.40
Benchmarks
How it scores.
- GPQA Diamond72.8%
82nd of 186
- FrontierMath36.1%
55th of 81
- ARC-AGI-26.1%
48th of 79
- Humanity's Last Exam14.0%
21st of 40
- SimpleQA Verified19.6%
66th of 77
- Mock AIME81.7%
78th of 176
Everything else Epoch has for it
- MATH level 597.8%
- Fiction.LiveBench77.8%
- Lech Mazur Writing75.0%
- Aider polyglot72.0%
- GeoBench64.0%
- METR Time Horizons63.9%
- DTBench62.7%
- CadEval62.0%
- ARC-AGI58.7%
- WeirdML52.6%
- VPCT36.3%
- LMCA31.2%
- SimpleBench26.4%
- GDPval25.3%
- Chess Puzzles22.1%
- FrontierMath-Tier-4-v2-Private4.9%
- GSO-Bench3.6%
- Mystery Game Puzzles0.0%
On the index
Around it.
Sources