o1
por OpenAI · Lançado em 2024-12-17
54.3
pontuação média
$15.00/1M
Preço de entrada
$60.00/1M
Preço de saída
200K tokens (~100 books)
Janela de contexto
multimodal
Tipo
Tested on 13 benchmarks with 54.3% average. Top scores: MATH level 5 (94.7%), Fiction.LiveBench (83.3%), GeoBench (80.0%).
Pontuações de benchmark
| Benchmark | Categoria | Pontuação | Bar |
|---|---|---|---|
| MATH level 5 | math | 94.7 | |
| Fiction.LiveBench | knowledge | 83.3 | |
| GeoBench | knowledge | 80.0 | |
| OTIS Mock AIME 2024-2025 | math | 73.3 | |
| Lech Mazur Writing | knowledge | 70.2 | |
| GPQA diamond | knowledge | 69.0 | |
| Aider polyglot | coding | 61.7 | |
| CadEval | coding | 56.0 | |
| WeirdML | coding | 43.8 | |
| ARC-AGI | reasoning | 30.7 | |
| SimpleBench | reasoning | 28.1 | |
| FrontierMath-2025-02-28-Private | math | 9.3 | |
| VPCT | knowledge | 5.5 |