Claude Opus 4.7 vs GPT-5.6 Sol
Claude Opus 4.7 
GPT-5.6 Sol Verdict
- GPT-5.6 Sol (high) leads on quality: 68.0 vs 65.5.
- Claude Opus 4.7 is stronger in human preference, maths, reasoning.
- GPT-5.6 Sol (high) is stronger in agents & tools, coding, composite, instruction following, knowledge, long context, multimodal.
- GPT-5.6 Sol (high) is 1.3× cheaper ($8.00 vs $10.00 per 1M blended).
| Metric | Claude Opus 4.7 | GPT-5.6 Sol (high) |
|---|---|---|
| BenchLeader Index | 65.5 | 68.0 |
| Agents & tools score | 69.4 | 87.6 |
| Coding score | 64.8 | 72.0 |
| Composite score | 80.3 | 82.5 |
| Human preference score | 68.8 | – |
| Instruction following score | 58.5 | 67.6 |
| Knowledge score | 66.4 | 73.7 |
| Long context score | 65.9 | 67.4 |
| Maths score | 65.2 | – |
| Multimodal score | 65.7 | 66.5 |
| Reasoning score | 66.3 | 62.9 |
| Blended price $/M | $10.00 | $8.00 |
| Output speed | 46 tok/s | 59 tok/s |
| Time to first answer | 21.2 s | 31.4 s |
| Context window | 1M | 1M |
| Humanity's Last Exam | 36.2% | – |
| Terminal-Bench | 80.2% | – |
| SimpleBench | 61.7% | – |
| SciCode | – | 56.9% |
| WeirdML | 76.4% | 88.8% |
| FrontierCode | 38.5% | – |
| GSO-Bench | 44.1% | – |
| Epoch Capabilities Index | 156.3 | – |
| LMArena Text | 1495 | – |
| LMArena Hard Prompts | 1519 | – |
| LMArena Coding | 1547 | – |
| LMArena WebDev | 1557 | – |
| LMArena Vision | 1317 | – |
| AA Intelligence Index | 40.7 | 42.5 |
| IFBench | 58.6% | 69.2% |
| AA-LCR | 78.7% | 81.7% |
| MMMU-Pro | 78.8% | 81.8% |
| AA-Omniscience | 27.3 | 20.4 |
| Terminal-Bench Hard | 51.5% | 62.1% |
| GPQA Diamond (AA) | 91.4% | 92.8% |
| Humanity's Last Exam (AA) | 42.3% | 46.0% |
| SciCode (AA) | – | 57.8% |
| τ²-Bench Telecom (AA) | 88.6% | 83.3% |
| AIME (Vals) | 96.3% | – |
| LiveCodeBench | 85.1% | – |
| MMLU-Pro | 89.9% | – |
| IOI | 47.1% | – |
| LegalBench | 85.3% | – |
| CorpFin | 66.1% | – |
| TaxEval | 75.3% | – |
| Terminal-Bench 2.1 (Vals) | 68.5% | – |
| SWE-bench (Vals) | 82.0% | – |
| GPQA Diamond (Vals) | 90.2% | – |
| Vals Index | 56.1 | – |
| EnigmaEval | – | 37.1% |
| HiL-Bench | 41.7% | – |
| EQ-Bench 4 | 1311 | – |
| Kagi LLM Benchmark | 73.3% | – |
| ARC-AGI-1 | – | 97.0% |
| ARC-AGI-2 | – | 85.4% |
| ARC-AGI-3 | – | 2.2% |
Data as of 2026-09-09. Best configuration of each model; every score links to its source on the model pages.