GPT-5.5 Pro (xHigh) — benchmark results
GPT-5.5 Pro evaluated at the xhigh reasoning-effort setting. Provider: OpenAI. Released 2026-04-23. Access: API.
Unified ELO 2268 ± 57, rank #2 of 1776 rated models, from 12 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Chess Puzzles (Epoch AI) | 64 | Accuracy (%) | 99.1 |
| OTIS Mock AIME 2024-25 | 100 | Accuracy (%) | 99 |
| Epoch AI - Critpt | 30.57 | Score | 98.6 |
| FrontierMath - Tiers 1-3 | 51 | Accuracy (%, 290 problems) | 98 |
| FrontierMath - Tier 4 | 39.6 | Accuracy (%, 48 problems) | 97.9 |
| Epoch AI - ECI | 160.93 | ECI Score | 97.8 |
| FrontierMath - Tiers 1-3 (v2) | 87.72 | Accuracy (%, 285 private v2 problems) | 97.3 |
| ARC-AGI-2 | 84.16 | Accuracy (%) | 96.3 |
| ARC-AGI-1 | 95 | Accuracy (%) | 95.3 |
| FrontierMath - Tier 4 (v2) | 78.05 | Accuracy (%, 41 private v2 problems) | 92.3 |
| IMO-Bench | 88.1 | Advanced ProofBench Accuracy (%) | 90.9 |
| SimpleQA Verified | 64.5 | Accuracy (%) | 87.5 |
Interactive version: theaggregate.ai/model?slug=gpt-5-5-pro-xhigh · How the rankings work · Data refreshed daily, snapshot 2026-07-22.