GPT-5.6 Sol (xHigh): benchmark results
GPT-5.6 Sol evaluated at the xhigh reasoning-effort setting. Provider: OpenAI. Released 2026-07-09. Access: API.
Aggregate Rating 1884 ± 9, rank #45 of 2168 rated models, from 285 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Chatbot Arena (Search) | 1257 | Arena Score | 100 |
| ComplexityWorld | 75.6 | Verifier acceptance rate (%; share of the 390 frozen main-pa | 100 |
| LMArena Search Arena | 1257.26 | Arena rating | 100 |
| LongDocBench - Caption Relationship Recovery | 65 | Caption relationship similarity (%; the model reads fixed Te | 100 |
| LongDocBench - Contextual Relationship Recovery | 63 | Macro similarity (%; unweighted mean of the caption, note an | 100 |
| LongDocBench - Note Relationship Recovery | 42 | Note relationship similarity (%; the model reads fixed TextI | 100 |
| Opus Magnum Bench | 70.86 | Human-normalized score (%) | 100 |
| PM-LLM-Benchmark | 40.2 | Score | 100 |
| PetriBench - Boundedness (Hard) | 69 | Exact-Match Accuracy (%) | 100 |
| PetriBench - Deadlock (Hard) | 92 | Exact-Match Accuracy (%) | 100 |
| PetriBench - Hard | 77.8 | Exact-Match Accuracy (%) | 100 |
| PetriBench - L0-Liveness (Hard) | 91 | Exact-Match Accuracy (%) | 100 |
Interactive version: theaggregate.ai/model?slug=gpt-5-6-sol-xhigh · How It Works · Data refreshed daily, snapshot 2026-10-20.