GPT-6 Sol: benchmark results
Provider: OpenAI. Access: API.
Unified ELO 2088 ± 17, rank #11 of 2928 rated models, from 15 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| BenchLM | 80.5 | Overall Score | 97.5 |
| RuneBench | 13372 | Total Peak XP Rate (XP/min) | 96.5 |
| LLM Stats (Agents' Last Exam) | 56.4 | Score (%) | 95 |
| LLM Stats Score | 49.39 | LLM Stats Score (conservative rating) | 94.6 |
| SvelteBench | 100 | Average pass@1 (%) | 94.6 |
| AI Chess Leaderboard (Continuation) | 1195 | Elo | 85.6 |
| AI Chess Leaderboard (Reasoning) | 1072 | Elo | 80.5 |
| LLM Stats (FrontierCode 1.1) | 49.3 | Score (%) | 78.9 |
| ParseBench | 68.19 | Overall Score | 77.8 |
| LLM Stats (DeepSWE 1.1) | 68.8 | Score (%) | 68.9 |
| LLM Stats (OSWorld 2.0) | 64.4 | Score (%) | 61.5 |
| TaxCalcBench | 26 | Correct Returns - Strict (%) | 53.6 |
Interactive version: theaggregate.ai/model?slug=gpt-6-sol · How It Works · Data refreshed daily, snapshot 2026-09-23.