GPT-5 Mini (Low): benchmark results
GPT-5 Mini evaluated at the low reasoning-effort setting. Provider: OpenAI. Released 2025-08-07. Access: API.
Unified ELO 1608 ± 1, rank #399 of 1761 rated models, from 13 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| SLR-Bench - Basic | 100 | Accuracy (%) | 86.4 |
| AJ-Bench | 72.41 | Overall Avg@3 (%) | 84.6 |
| FormationEval | 95.2 | Accuracy (%) | 78.2 |
| SLR-Bench - Hard | 8 | Accuracy (%) | 70 |
| SLR-Bench - Medium | 26 | Accuracy (%) | 67.3 |
| SLR-Bench - Easy | 85 | Accuracy (%) | 65.5 |
| IH-Benchmark | 85.8 | Overall Compliance (%) | 61.1 |
| Chess Puzzles (Epoch AI) | 12 | Accuracy (%) | 43.6 |
| LLM Chess (Saplin) | -55.6 | ELO | 43.5 |
| FrontierMath - Tiers 1-3 (v2) | 18.25 | Accuracy (%, 285 private v2 problems) | 24.4 |
| ARC-AGI-1 | 26.33 | Accuracy (%) | 22.3 |
| MCPMark | 8.27 | Pass@1 (%) | 18.4 |
Interactive version: theaggregate.ai/model?slug=gpt-5-mini-low · How It Works · Data refreshed daily, snapshot 2026-09-05.