GPT-5.5 Pro: benchmark results
OpenAI's higher-compute GPT-5.5 variant for demanding reasoning and general tasks. Provider: OpenAI. Released 2026-04-23. Access: API.
Aggregate Rating 1899 ± 13, rank #17 of 1656 rated models, from 76 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| FrontierMath - Tiers 1-3 | 52.4 | Accuracy (%, 290 problems) | 100 |
| GeneBench | 33.2 | Mean pass rate (imported) | 100 |
| OpenAI GPT-5.5 Launch - BrowseComp | 90.1 | Score (%) | 100 |
| OpenAI GPT-5.5 Launch - FrontierMath Tier 4 | 39.6 | Score (%) | 100 |
| OpenAI GPT-5.5 Launch - GeneBench | 33.2 | Score (%) | 100 |
| OpenAI GPT-5.5 Launch - Investment Banking Modeling Tasks | 88.6 | Score (%) | 100 |
| OpenAI GPT-5.5 System Card - Biochemistry Knowledge Improvement | 39.26 | reward@4 (%) | 100 |
| OpenAI GPT-5.5 System Card - DNA Sequence Design for TF Binding | 16.5 | pass@1 (%) | 100 |
| OpenAI GPT-5.6 System Card - DNA Sequence Design for TF Binding | 16.5 | pass@1 (%) | 100 |
| OpenAI GPT-Rosalind-5.5 System Card - Biorisk Knowledge | 81.67 | cons@32 (%) | 100 |
| OpenAI GPT-Rosalind-5.5 System Card - DNA Sequence Design for TF Binding | 16.5 | pass@1 (%) | 100 |
| OpenAI GPT-Rosalind-5.5 System Card - Multi-select Virology Troubleshooting | 55.34 | pass@1 (%) | 100 |
Interactive version: theaggregate.ai/model?slug=gpt-5-5-pro · How It Works · Data refreshed daily, snapshot 2026-10-20.