OpenAI
GPT-5.4 Mini benchmark results
9 results on 5 benchmarks from 2 sources.
Also listed as: gpt-5.4-mini-high · gpt-5.4-mini-2026-03-17_xhigh · GPT-5.4 mini (xhigh) · gpt-5.4-mini-2026-03-17_none · GPT-5.4 mini (none) · gpt-5.4-mini-2026-03-17_high · GPT-5.4 mini (high)
All benchmark results
| Benchmarks | Score | Rank | Setting | Source type | Date |
|---|---|---|---|---|---|
| LMArena Text | 1447 | 84 | Overall, style control on; 63493 votes; 95% CI 1443–1451“gpt-5.4-mini-high” | Official leaderboard | Oct 8, 2026 |
| LMArena WebDev | 1397 | 92 | Overall; 11976 votes; 95% CI 1391–1404“gpt-5.4-mini-high” | Official leaderboard | Oct 8, 2026 |
| LMArena Vision | 1251 | 50 | Overall, style control on; 25734 votes; 95% CI 1244–1257“gpt-5.4-mini-high” | Official leaderboard | Oct 2, 2026 |
| GPQA Diamond | 86.9% | 73 | Epoch AI run, accuracy; effort: xhigh | Independent run | Aug 7, 2026 |
| GPQA Diamond | 83.6% | 94 | Epoch AI run, accuracy; effort: high | Independent run | Apr 15, 2026 |
| GPQA Diamond | 64.1% | 177 | Epoch AI run, accuracy; effort: none | Independent run | Aug 7, 2026 |
| AIME (OTIS Mock AIME 2024–2025) | 88.9% | 73 | Epoch AI run, OTIS Mock AIME 2024–2025; effort: xhigh | Independent run | Aug 7, 2026 |
| AIME (OTIS Mock AIME 2024–2025) | 87.2% | 82 | Epoch AI run, OTIS Mock AIME 2024–2025; effort: high | Independent run | Apr 15, 2026 |
| AIME (OTIS Mock AIME 2024–2025) | 26.7% | 197 | Epoch AI run, OTIS Mock AIME 2024–2025; effort: none | Independent run | Aug 7, 2026 |
FAQ
Where do these GPT-5.4 Mini 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.