Up to four models
GPT-5.6 Sol vs GPT-5.5
On the index (GPT-4 = 100), GPT-5.6 Sol scores 179 and GPT-5.5 172. Side by side on the same third-party tests, with price and context. The link in your address bar shares this exact comparison.
GPT-5.6 Sol
179164–195
WeightsClosed
ReleasedJul 2026
Price in / out$4.0 / $20
Context1.05M
Core tests taken5 of 9
GPT-5.5
172156–188
WeightsClosed
ReleasedApr 2026
Price in / out$5.0 / $30
Context1.05M
Core tests taken6 of 9
GPT-5.6 Sol leads GPT-5.5 by 7 points, ahead on 3 of 4 shared core tests, and costs 1.4× less.
Test by testthird-party results
GPT-5.6 SolGPT-5.5
Every result as a table
| Test | GPT-5.6 Sol | GPT-5.5 |
|---|---|---|
| Reasoning & knowledge | ||
| GPQA Diamond (Epoch AI) | 93.5% | 94.0% |
| LiveBench Data Analysis | 79.8% | 81.6% |
| LiveBench IF | 71.9% | 70.7% |
| LiveBench Language | 87.7% | 87.4% |
| LiveBench Reasoning | 91.6% | 89.6% |
| GPQA Diamond (Vals AI) | 95.2% | 93.2% |
| MMLU Pro | 89.1% | 88.1% |
| MMMU Pro | 88.8% | 88.3% |
| EBR-bench | 44.8% | 34.3% |
| SimpleBench | 64.8% | 69.0% |
| SimpleQA Verified | 69.7% | 63.0% |
| Maths | ||
| FrontierMath (tiers 1-3) | 89.1% | 85.3% |
| AIME-style maths (OTIS mock) | 100.0% | 100.0% |
| LiveBench Mathematics | 96.2% | 95.9% |
| Chess puzzles | 59.0% | 63.0% |
| FrontierMath (tier 4) | 82.9% | 72.5% |
| Surface Evolver Bench | 93.1% | 88.1% |
| Coding | ||
| LiveBench Agentic Coding | 56.2% | 54.0% |
| LiveBench Coding | 83.9% | 82.2% |
| Code Migration | 52.9% | 45.2% |
| LiveCodeBench | 82.6% | 85.3% |
| ProgramBench | 1.5% | 0.5% |
| SWE-bench | 96.2% | 82.6% |
| Vibe Code Bench v1.1 | 80.5% | 69.8% |
| DeepSWE | 72.7% | 67.0% |
| GSO-Bench | 76.5% | 40.2% |
| MirrorCode | 20.0% | 23.3% |
| WeirdML | 88.8% | 84.9% |
| Agents | ||
| SkillsBench | 54.1% | 62.2% |
| SRE Bench | 30.5% | 3.8% |
| DTBench | 95.7% | 96.0% |
| Furniture assembly | 56.7% | 44.2% |
| LMCA | 58.5% | 54.3% |
| Mystery game puzzles | 58.0% | 56.0% |
| Professional work | ||
| Finance Agent (v2) | 53.8% | 51.8% |
| CorpFin v2 | 64.4% | 68.4% |
| EMB | 72.3% | 64.5% |
| LegalBench | 87.0% | 86.5% |
| Legal Research Bench | 48.1% | 40.4% |
| MedCode | 44.0% | 49.1% |
| MedScribe | 85.2% | 86.9% |
| MortgageTax | 67.3% | 68.8% |
| Public Benefits Bench v1.1 | 66.5% | 60.9% |
| Tax Agent Bench | 31.1% | 21.8% |
| TaxEval v2 | 74.8% | 75.0% |
| Novel problems | ||
| ARC-AGI-2 | 92.5% | 85.0% |
| ARC-AGI-1 | 97.5% | 95.0% |
| ARC-AGI-3 | 7.8% | 0.4% |
Tests at least two of these models have taken. Bold rows are the core tests.