GPT-4o mini
#178 of 274 scored.
OpenAIJul 18, 2024Closed
Capabilities Index
Range 119.1–128.7
126.6
Context
Takes text, image, file
128K
Input
Per 1M tokens
$0.15
Output
Per 1M tokens
$0.60
Benchmarks
How it scores.
- GPQA Diamond17.0%
162nd of 186
- FrontierMath0.7%
78th of 81
- ARC-AGI-20.0%
72nd of 79
- SimpleQA Verified8.3%
76th of 77
- Mock AIME6.9%
143rd of 176
Everything else Epoch has for it
- GSM8K91.3%
- PIQA77.4%
- MMLU75.7%
- Lech Mazur Writing67.2%
- GeoBench64.0%
- MATH level 552.6%
- DTBench24.0%
- Balrog17.4%
- LMCA12.2%
- WeirdML11.8%
- Aider polyglot3.6%
- Mystery Game Puzzles3.1%
- VPCT1.0%
- Chess Puzzles0.0%
- SimpleBench0.0%
On the index
Around it.
Sources