GPT-5.2: benchmark results
OpenAI's standard GPT-5.2 model for general-purpose reasoning, coding, and writing. Provider: OpenAI. Released 2025-12-11. Access: API.
Unified ELO 1681 ± 1, rank #82 of 1392 rated models, from 969 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| ALL Bench Multimodal | 86.7 | Average Numeric VLM Score (%) | 100 |
| AMA-Bench | 69.83 | Average Score (%) | 100 |
| AMA-Bench - Game | 81.13 | Average Score (%) | 100 |
| CLEM ImageGame | 99.92 | Game Clemscore (%) | 100 |
| CringeBench | 0.29 | Cringe Score (0-10, lower is better) | 100 |
| DramaBench | 96.04 | Overall Score (%) | 100 |
| EHRBench | 70.91 | Overall Acc (%) (self-reported) | 100 |
| EsoLang-Bench | 4.2 | Accuracy (%) | 100 |
| Factory Code Review Benchmark | 60.5 | Mean F1 (%) | 100 |
| From Perception to Action | 22.9 | pass@1 (self-reported) | 100 |
| GDPval-MM | 70.9 | Wins + ties (self-reported) | 100 |
| GPQA Diamond (Opus 4.6 System Card) | 93.2 | Accuracy (%) | 100 |
Interactive version: theaggregate.ai/model?slug=gpt-5-2 · How It Works · Data refreshed daily, snapshot 2026-09-05.