OpenAI
GPT-5 nano benchmark results
24 results on 8 benchmarks from 4 sources.
Also listed as: gpt-5-nano-high · gpt-5-nano-2025-08-07_low · GPT-5 nano (low) · gpt-5-nano-2025-08-07_minimal · GPT-5 nano (minimal) · gpt-5-nano-2025-08-07_high · GPT-5 nano (high) · gpt-5-nano-2025-08-07_medium · GPT-5 nano (medium) · gpt-5-nano-2025-08-07_unknown
All benchmark results
| Benchmarks | Score | Rank | Setting | Source type | Date |
|---|---|---|---|---|---|
| LMArena Text | 1337 | 234 | Overall, style control on; 8140 votes; 95% CI 1330–1344“gpt-5-nano-high” | Official leaderboard | Oct 8, 2026 |
| LMArena Vision | 1146 | 112 | Overall, style control on; 3930 votes; 95% CI 1135–1158“gpt-5-nano-high” | Official leaderboard | Oct 2, 2026 |
| GPQA Diamond | 69.4% | 163 | Epoch AI run, accuracy; effort: high | Independent run | Oct 30, 2025 |
| GPQA Diamond | 67.4% | 167 | Epoch AI run, accuracy; effort: medium | Independent run | Aug 7, 2025 |
| GPQA Diamond | 57.6% | 185 | Epoch AI run, accuracy; effort: low | Independent run | Jul 13, 2026 |
| GPQA Diamond | 48.5% | 211 | Epoch AI run, accuracy; effort: minimal | Independent run | Jul 13, 2026 |
| SWE-bench Verified | 34.8% | 88 | agent: mini-SWE-agent“gpt-5-nano-2025-08-07” | Official leaderboard | Aug 7, 2025 |
| AIME (OTIS Mock AIME 2024–2025) | 81.1% | 109 | Epoch AI run, OTIS Mock AIME 2024–2025; effort: high | Independent run | Oct 31, 2025 |
| AIME (OTIS Mock AIME 2024–2025) | 74.2% | 119 | Epoch AI run, OTIS Mock AIME 2024–2025; effort: medium | Independent run | Aug 7, 2025 |
| AIME (OTIS Mock AIME 2024–2025) | 46.7% | 172 | Epoch AI run, OTIS Mock AIME 2024–2025; effort: low | Independent run | Jul 13, 2026 |
| AIME (OTIS Mock AIME 2024–2025) | 35.6% | 187 | Epoch AI run, OTIS Mock AIME 2024–2025; effort: minimal | Independent run | Jul 13, 2026 |
| MATH Level 5 | 95.2% | 14 | Epoch AI run, accuracy; effort: medium | Independent run | Aug 20, 2025 |
| MATH Level 5 | 94.9% | 16 | Epoch AI run, accuracy; effort: high | Independent run | Aug 20, 2025 |
| Terminal-Bench 2.0 | 21.8% | 164 | Terminal-Bench 2.0; agent: spoox-o-m; effort: unknown thinking | Official leaderboard via Epoch AI | Aug 7, 2025 |
| Terminal-Bench 2.0 | 11.5% | 185 | Terminal-Bench 2.0; agent: Codex CLI ; effort: medium | Official leaderboard via Epoch AI | Nov 4, 2025 |
| Terminal-Bench 2.0 | 11.5% | 186 | Terminal-Bench 2.0; agent: Codex CLI; effort: unknown thinking | Official leaderboard via Epoch AI | Aug 7, 2025 |
| Terminal-Bench 2.0 | 9.9% | 187 | Terminal-Bench 2.0; agent: OpenHands ; effort: medium | Official leaderboard via Epoch AI | Nov 2, 2025 |
| Terminal-Bench 2.0 | 9.9% | 188 | Terminal-Bench 2.0; agent: OpenHands; effort: unknown thinking | Official leaderboard via Epoch AI | Aug 7, 2025 |
| Terminal-Bench 2.0 | 7.9% | 190 | Terminal-Bench 2.0; agent: Terminus 2 ; effort: medium | Official leaderboard via Epoch AI | Oct 31, 2025 |
| Terminal-Bench 2.0 | 7.9% | 191 | Terminal-Bench 2.0; agent: Terminus 2; effort: unknown thinking | Official leaderboard via Epoch AI | Aug 7, 2025 |
| Terminal-Bench 2.0 | 7.0% | 192 | Terminal-Bench 2.0; agent: Mini-SWE-Agent ; effort: medium | Official leaderboard via Epoch AI | Nov 3, 2025 |
| Terminal-Bench 2.0 | 7.0% | 193 | Terminal-Bench 2.0; agent: Mini-SWE-Agent; effort: unknown thinking | Official leaderboard via Epoch AI | Aug 7, 2025 |
| BFCL v4 (function calling) | 51.5% | 24 | Overall accuracy; native function calling (FC)“GPT-5-nano-2025-08-07 (FC)” | Official leaderboard | Oct 17, 2026 |
| BFCL v4 (function calling) | 27.6% | 79 | Overall accuracy; prompt-based“GPT-5-nano-2025-08-07 (Prompt)” | Official leaderboard | Oct 17, 2026 |
FAQ
Where do these GPT-5 nano scores come from?
From Epoch AI Benchmarking Hub, Berkeley Function Calling Leaderboard (BFCL), LMArena leaderboard dataset, and SWE-bench leaderboards (swebench.com). Each row links to the original leaderboard, dataset or report, and vendor-reported numbers are labelled.