OpenAI
GPT-4 (Mar 2023) benchmark results
4 results on 4 benchmarks from 2 sources.
Also listed as: gpt-4-0314 · GPT-4 (initial release)
All benchmark results
| Benchmarks | Score | Rank | Setting | Source type | Date |
|---|---|---|---|---|---|
| LMArena Text | 1288 | 294 | Overall, style control on; 54173 votes; 95% CI 1283–1293“gpt-4-0314” | Official leaderboard | Oct 8, 2026 |
| GPQA Diamond | 35.7% | 247 | Epoch AI run, accuracy; effort: Mar 2023 | Independent run | Oct 23, 2025 |
| MMLU | 86.4% | 3 | Exact match; 5-shot; effort: Mar 2023 | Paper / tech report via Epoch AI | Mar 14, 2023 |
| AIME (OTIS Mock AIME 2024–2025) | 0.6% | 260 | Epoch AI run, OTIS Mock AIME 2024–2025; effort: Mar 2023 | Independent run | Oct 22, 2025 |
FAQ
Where do these GPT-4 (Mar 2023) scores come from?
From Epoch AI Benchmarking Hub and LMArena leaderboard dataset. Each row links to the original leaderboard, dataset or report, and vendor-reported numbers are labelled.