GPT-5.2 (Low): benchmark results
GPT-5.2 evaluated at the low reasoning-effort setting. Provider: OpenAI. Released 2025-12-11. Access: API.
Aggregate Rating 1678 ± 14, rank #427 of 2168 rated models, from 63 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| CoopEval - No Mechanism | 0.14 | Normalized mean payoff (open scale; below 0 when exploited, | 100 |
| Med-CoReasoner - Global-MMLU-Medical | 84.92 | Accuracy (%) | 100 |
| Med-CoReasoner - Global-MMLU-Medical - Chinese | 84.39 | Accuracy (%) | 100 |
| Med-CoReasoner - Global-MMLU-Medical - Italian | 85.78 | Accuracy (%) | 100 |
| Med-CoReasoner - Global-MMLU-Medical - Japanese | 86.25 | Accuracy (%) | 100 |
| Med-CoReasoner - Global-MMLU-Medical - Korean | 83.65 | Accuracy (%) | 100 |
| Med-CoReasoner - Global-MMLU-Medical - Spanish | 85.98 | Accuracy (%) | 100 |
| Med-CoReasoner - Global-MMLU-Medical - Swahili | 81.86 | Accuracy (%) | 100 |
| Med-CoReasoner - MMLU-ProX-Health | 74.53 | Accuracy (%) | 100 |
| Med-CoReasoner - MMLU-ProX-Health - Chinese | 72.63 | Accuracy (%) | 100 |
| Med-CoReasoner - MMLU-ProX-Health - French | 77 | Accuracy (%) | 100 |
| Med-CoReasoner - MMLU-ProX-Health - German | 76.42 | Accuracy (%) | 100 |
Interactive version: theaggregate.ai/model?slug=gpt-5-2-low · How It Works · Data refreshed daily, snapshot 2026-10-20.