GPT-5.2 Pro: benchmark results
OpenAI's higher-compute GPT-5.2 variant for difficult reasoning and general tasks. Provider: OpenAI. Released 2025-12-11. Access: API.
Unified ELO 1713 ± 1, rank #38 of 1392 rated models, from 62 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| AutoBench | 4.48 | Judge Rating (1-5) | 100 |
| LLM Stats (HMMT 2025) | 100 | Score (%) | 100 |
| SuperGPQA | 67.13 | Accuracy (%) | 98.5 |
| ZeroEval GPQA Diamond | 93.2 | GPQA Diamond Score | 96.7 |
| Pencil Puzzle Bench - Sudoku | 6.7 | Direct-ask Success Rate (%) | 96 |
| LM Market Cap LMC Score | 90.5 | LMC Score (0-100) | 94.5 |
| IUMB | 89.6 | Score (%) | 94.4 |
| SGI-Bench Idea Generation | 55.03 | Idea Generation Score | 94.4 |
| FrontierMath - Tier 4 | 31.3 | Accuracy (%, 48 problems) | 93 |
| Pencil Puzzle Bench - Norinori | 53.3 | Direct-ask Success Rate (%) | 91 |
| Pencil Puzzle Bench - Nurimisaki | 20 | Direct-ask Success Rate (%) | 91 |
| DEEPSYNTH | 8.7 | F1 (self-reported) | 90.9 |
Interactive version: theaggregate.ai/model?slug=gpt-5-2-pro · How It Works · Data refreshed daily, snapshot 2026-09-05.