OpenAIGPT
GPT-5.6 Sol
OpenAI · GPT. Released Jul 9, 2026.
GPT-5.6 Sol is a model from OpenAI in the GPT family, released Jul 9, 2026. evals.report tracks 10 reported GPT-5.6 Sol benchmark scores across DeepSWE, SWE-Marathon, FrontierCode, Terminal-Bench 2.1, SWE-bench Pro, GPQA Diamond, BrowseComp, MMMU-Pro, and 2 more — each shown with its benchmark, metric, source status, and date, and never combined into a single ranking.
10 results
Benchmark results 10
Compare this model| Benchmark | Category | Score | Metric | Status | Date | |
|---|---|---|---|---|---|---|
| DeepSWE | Coding | 73.0% | % resolved | Unverified | Jul 9, 2026 | Details |
| SWE-Marathon | Agents | 39.0% | resolution rate (pass@1) | Unverified | Jul 9, 2026 | Details |
| FrontierCode | Coding | 47.5% | weighted score (Main) | Official | Jul 9, 2026 | Details |
| Terminal-Bench 2.1 | Agents | 88.8% | task success | Verified | Jul 9, 2026 | Details |
| SWE-bench Pro | Coding | 64.6% | % resolved | Verified | Jul 9, 2026 | Details |
| GPQA Diamond | Reasoning | 94.6% | accuracy | Verified | Jul 9, 2026 | Details |
| BrowseComp | Agents | 90.4% | accuracy | Verified | Jul 9, 2026 | Details |
| MMMU-Pro | Multimodal | 83.0% | accuracy | Verified | Jul 9, 2026 | Details |
| OpenAI-MRCR v2 (Multi-Round Coreference Resolution) | Reasoning | 73.8% | accuracy (mean SequenceMatcher similarity) | Verified | Jul 9, 2026 | Details |
| ARC-AGI-3 | Reasoning | 7.78% | accuracy | Verified | Jul 9, 2026 | Details |