GPT-4.1
#133 of 274 scored.
OpenAIApr 14, 2025Closed
Capabilities Index
Range 133.6–138.6
136.8
Context
Takes image, text, file
1.05M
Input
Per 1M tokens
$2.00
Output
Per 1M tokens
$8.00
Benchmarks
How it scores.
- GPQA Diamond55.9%
106th of 186
- SWE-bench Verified48.5%
31st of 32
- FrontierMath6.0%
77th of 81
- ARC-AGI-20.4%
70th of 79
- Humanity's Last Exam0.6%
35th of 40
- SimpleQA Verified31.1%
58th of 77
- Mock AIME38.3%
116th of 176
Everything else Epoch has for it
- MATH level 583.0%
- GeoBench72.0%
- Fiction.LiveBench63.9%
- Aider polyglot52.4%
- DTBench47.1%
- CadEval42.0%
- WeirdML39.0%
- LMCA30.1%
- SimpleBench12.4%
- ARC-AGI5.5%
- Chess Puzzles1.1%
- FrontierMath-Tier-4-2025-07-01-Private0.0%
On the index
Around it.
Sources