GPT-5.6 Sol (High): benchmark results
GPT-5.6 Sol evaluated at the high reasoning-effort setting. Provider: OpenAI. Released 2026-07-09. Access: API.
Aggregate Rating 1885 ± 8, rank #44 of 2168 rated models, from 260 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| AeroCopilotBench - Tier-2 Safety-Gated Outcome | 91.1 | Goal attainment (%; zero on any safety violation) | 100 |
| AeroCopilotBench - Tier-2 Safety-Gated Success Rate | 72.6 | Success rate (%; goals met with no safety violation) | 100 |
| BoundaryBench (Unrestricted) | 83.9 | Success Rate (%) | 100 |
| Grid-Bench T1 D2 exact | 62 | Exact answers (out of 64) | 100 |
| Grid-Bench T1 D3 exact | 60 | Exact answers (out of 64) | 100 |
| Grid-Bench T1 D4 exact | 61 | Exact answers (out of 64) | 100 |
| Grid-Bench T1 pooled exact | 183 | Exact answers (out of 192) | 100 |
| Grid-Bench T16 D2 exact | 49 | Exact answers (out of 64) | 100 |
| Grid-Bench T16 D3 exact | 21 | Exact answers (out of 64) | 100 |
| Grid-Bench T16 D4 exact | 17 | Exact answers (out of 64) | 100 |
| Grid-Bench T16 pooled exact | 87 | Exact answers (out of 192) | 100 |
| Grid-Bench T32 pooled exact | 57 | Exact answers (out of 192) | 100 |
Interactive version: theaggregate.ai/model?slug=gpt-5-6-sol-high · How It Works · Data refreshed daily, snapshot 2026-10-20.