GPT-5-Nano is the smallest and fastest variant in the GPT-5 system, optimized for developer tools, rapid interactions, and ultra-low latency environments. While limited in reasoning depth compared to its larger... (description from the OpenRouter listing)
| Benchmark | Domain | Score | vs best recorded | Setting | Run | Source |
|---|---|---|---|---|---|---|
| LMCA | agents | 7.9% | 12% | high | — | External ↗ |
| DTBench | reasoning | 62.7% | 64% | high | — | External ↗ |
| Mystery Game Puzzles | games | 8.0% ±2.7 | 10% | high | 27 Aug 2026 | Epoch ↗ |
| FrontierMath-Tiers-1-3-v2-Private | math | 20.0% ±2.4 | 21% | high | 12 Jun 2026 | Eval log ↗ |
| FrontierMath-Tier-4-v2-Private | math | 2.4% ±2.4 | 2% | high | 12 Jun 2026 | Eval log ↗ |
| ProofBench | math | 12.0% | 12% | high | — | External ↗ |
| Chess Puzzles | games | 27.0% ±4.5 | 38% | high | 7 Aug 2026 | Eval log ↗ |
| SimpleQA Verified | knowledge | 11.7% ±1.0 | 15% | high | 10 Aug 2026 | Epoch ↗ |
| FrontierMath-Tier-4-2025-07-01-Privatesuperseded | math | 0.0% | 0% | high | 30 Oct 2025 | Epoch ↗ |
| ARC-AGI-2 | reasoning | 2.6% | 3% | high | — | External ↗ |
| FrontierMath-2025-02-28-Privatesuperseded | math | 8.3% ±1.6 | 16% | high | 30 Oct 2025 | Epoch ↗ |
| VPCT | multimodal | 37.2% | 41% | high | — | External ↗ |
| WeirdML | coding | 38.1% ±0.0 | 41% | high | — | External ↗ |
| OTIS Mock AIME 2024-2025 | math | 81.1% ±4.6 | 81% | high | 31 Oct 2025 | Eval log ↗ |
| GPQA diamond | science | 69.4% ±2.8 | 73% | high | 30 Oct 2025 | Eval log ↗ |
| MATH level 5 | math | 94.9% ±0.5 | 97% | high | 20 Aug 2025 | Epoch ↗ |
| ARC-AGI | reasoning | 16.7% | 17% | high | — | External ↗ |
Source: Epoch AI Benchmarking Hub (CC BY 4.0). “External” rows are leaderboard results Epoch collects from third parties. Best reported setting per benchmark is shown.
13 observations (OpenRouter listing + Internet Archive snapshots). History accumulates with every ingest run; a single point means no change has been observed yet.