GPT-4o (2024-05-13): benchmark results
May 13, 2024 GPT-4o snapshot, tracked when sources report the dated API model. Provider: OpenAI. Released 2024-05-13. Access: API.
Aggregate Rating 1642 ± 6, rank #325 of 1656 rated models, from 1164 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| BankMathBench (English) - Intermediate | 14.4 | Exact-match accuracy (%) on the 423 English intermediate (co | 100 |
| BenchBench | 97.67 | Aggregate Score (%) | 100 |
| BigCodeBench | 51.1 | Pass@1 (%) | 100 |
| HELM (Stanford) | 93.76 | Mean Win Rate (%) | 100 |
| HELM Finance - BANKING77 | 78.5 | EM | 100 |
| HELM Lite | 95.95 | Mean win rate (imported) | 100 |
| HELM MedHELM - ADHD-Behavior | 83.69 | EM | 100 |
| HELM MedHELM - EHRSQL | 32 | EHRSQLExeAcc | 100 |
| HELM MedHELM - SHC-ENT | 61.1 | EM | 100 |
| HELM NarrativeQA | 80.39 | F1 (%) | 100 |
| HELM VHELM - A Okvqa Robustness | 89.1 | EM (Robustness) | 100 |
| HELM VHELM - Mmmu - Diagnostics and Laboratory Medicine | 60 | Prefix Quasi-Exact Match (%) | 100 |
Interactive version: theaggregate.ai/model?slug=gpt-4o-2024-05-13 · How It Works · Data refreshed daily, snapshot 2026-10-20.