GPT-5 Nano vs o4 Mini
Lado a lado. Cada métrica. Cada benchmark.
| Tipo | GPT-5 Nano | o4 Mini |
|---|---|---|
| Provider |  OpenAI | OpenAI |
| pontuação média | 45.3 | 53.2 |
| Preço de entrada | $0.05 | $1.10 |
| Preço de saída | $0.40 | $4.40 |
| Janela de contexto | 400K tokens (~200 books) | 200K tokens (~100 books) |
| Lançado em | 2025-08-07 | 2025-04-16 |
| Código aberto | Proprietary | Proprietary |
Pontuações de benchmark
17 benchmarks · GPT-5 Nano: 1, o4 Mini: 16
| Benchmark | Categoria | GPT-5 Nano | o4 Mini |
|---|---|---|---|
| ARC-AGI | reasoning | 20.7 | 58.7 |
| ARC-AGI-2 | reasoning | 2.6 | 6.1 |
| Fiction.LiveBench | knowledge | 44.4 | 77.8 |
| FrontierMath-2025-02-28-Private | math | 8.3 | 24.8 |
| FrontierMath-Tier-4-2025-07-01-Private | math | 2.1 | 6.3 |
| GPQA diamond | knowledge | 59.3 | 72.8 |
| HELM — GPQA | knowledge | 67.9 | 73.5 |
| HELM — IFEval | language | 93.2 | 92.9 |
| HELM — MMLU-Pro | knowledge | 77.8 | 82.0 |
| HELM — Omni-MATH | math | 54.7 | 72.0 |
| HELM — WildBench | reasoning | 80.6 | 85.4 |
| MATH level 5 | math | 95.2 | 97.8 |
| OTIS Mock AIME 2024-2025 | math | 81.1 | 81.7 |
| SimpleQA Verified | knowledge | 12.2 | 23.9 |
| SWE-Bench Verified (Bash Only) | coding | 34.8 | 45.0 |
| VPCT | knowledge | 5.8 | 36.3 |
| WeirdML | coding | 38.1 | 52.6 |