GPT-5.6 Pro Terra: benchmark results
Provider: OpenAI. Released 2026-07-09. Access: API.
Aggregate Rating 1855 ± 45, rank #32 of 1635 rated models, from 14 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| FlavourBench - Pairing | 72.11 | Pairing Family Score (0-100) | 92.3 |
| Tinybird AI SQL Benchmark - Exactness | 55.92 | Result exactness vs human reference queries (0-100) | 91.4 |
| LM Market Cap LMC Score | 89 | LMC Score (0-100) | 85.7 |
| FlavourBench | 63.67 | FlavourBench Score (0-100) | 84.6 |
| Tinybird AI SQL Benchmark - First-Attempt Success Rate | 98 | Questions answered with a valid query on the first attempt ( | 80.9 |
| FlavourBench - Constraint | 52.82 | Constraint Family Score (0-100) | 80.8 |
| AI Chess Leaderboard (Reasoning) | 1041 | Elo | 75 |
| Tinybird AI SQL Benchmark - Success Rate | 100 | Questions answered with a valid query within 3 retries (%) | 71.9 |
| FlavourBench - Substitution | 66.07 | Substitution Family Score (0-100) | 69.2 |
| PointerBench - Pro | 74 | Success rate (%) | 63.6 |
| PointerBench - Average | 48.7 | Success rate (%) | 59.1 |
| PointerBench - Text | 20 | Success rate (%) | 56.8 |
Interactive version: theaggregate.ai/model?slug=gpt-5-6-pro-terra · How It Works · Data refreshed daily, snapshot 2026-11-05.