# Claude 3.7 Sonnet vs GPT-5.6 Sol

> GPT-5.6 Sol is the stronger model overall, scoring 64.4 to 39.4 on the Noometry Index.

- Canonical page: https://noometry.com/compare/claude-3-7-sonnet-vs-gpt-5-6-sol
- Last updated: 2026-10-10
- Shared benchmarks: 26

## Summary

- They share 26 benchmarks with published results for both. Claude 3.7 Sonnet scores higher in 1 category and GPT-5.6 Sol in 9 categories; 10 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where GPT-5.6 Sol leads 74.7 to 18.7.
- The biggest single-benchmark swing is ARC-AGI-2: 0.9% for Claude 3.7 Sonnet and 92.5% for GPT-5.6 Sol.

## Snapshot

| | Claude 3.7 Sonnet | GPT-5.6 Sol |
|---|---|---|
| Provider | Anthropic | OpenAI |
| Noometry Index | 39.4 | 64.4 |
| Rank | 154 | 6 |
| Context | — | 1.05M |
| Input $/M | — | $4 |
| Output $/M | — | $20 |
| Weights | Proprietary | Proprietary |

## Coding

- Claude 3.7 Sonnet: 41.1 (#120)
- GPT-5.6 Sol: 61.1 (#7)

| Benchmark | Claude 3.7 Sonnet | GPT-5.6 Sol |
|---|---|---|
| LMArena Coding | 1361 | 1493 |
| SWE-bench Verified | 61% | — |
| DeepSWE | — | 72.7% |
| FrontierCode | — | 47.5% |
| SWE-bench Verified (bash only) | 52.8% | — |
| Aider Polyglot | 64.9% | — |
| CursorBench | — | 41.7% |
| LMArena WebDev | — | 1619 |
| FrontierSWE | — | 32.2% |
| SciCode | — | 57.1% |
| GSO | 3.8% | — |
| WeirdML | — | 89.4% |
| LiveBench Coding | 74.5% | — |
| MirrorCode | — | 20% |
| CadEval | 54% | — |
| ALE-Bench | — | 2,177 |

## Agentic & Tool Use

- Claude 3.7 Sonnet: 33.9 (#46)
- GPT-5.6 Sol: 50.2 (#7)

| Benchmark | Claude 3.7 Sonnet | GPT-5.6 Sol |
|---|---|---|
| APEX-Agents | — | 51.4% |
| OSWorld 2.0 | — | 27.3% |
| TheAgentCompany | 30.9% | — |
| τ²-bench Banking | — | 46.9% |
| Cybench | 20% | — |
| DeepResearch Bench | 43.6% | — |
| OSWorld | 35.8% | — |
| PostTrainBench | — | 36.2% |
| BALROG | — | 60% |
| GBAEval | — | 52.6% |
| GDP.pdf | — | 30.7% |
| LMArena Search | — | 1257 |
| METR Time Horizons | 60% | — |
| Vending-Bench 2 | — | 9,619 |

## Reasoning

- Claude 3.7 Sonnet: 18.7 (#252)
- GPT-5.6 Sol: 74.7 (#6)

| Benchmark | Claude 3.7 Sonnet | GPT-5.6 Sol |
|---|---|---|
| ARC-AGI-2 | 0.9% | 92.5% |
| SimpleBench | 46.4% | 71.7% |
| ARC-AGI-1 | 28.6% | 97.5% |
| EnigmaEval | 4.2% | 37.1% |
| LMArena Hard Prompts | 1333 | 1481 |
| Epoch Capabilities Index | 141.16 | 161.99 |
| Kagi LLM Benchmark | — | 67% |
| NYT Connections (extended) | — | 93.8% |
| CritPt | — | 32.3% |
| Chess Puzzles | — | 64% |
| EBR-Bench | — | 44.8% |
| LiveBench Reasoning | 87.8% | — |
| Mystery Game Puzzles | — | 58% |
| DTBench | — | 96% |
| LiveBench Data Analysis | 74% | — |
| LMCA | — | 59.2% |
| Surface Evolver Bench | — | 93.1% |
| Bench to the Future 3 | — | 0.14 |
| ForecastBench | 61.8 | — |
| LiveBench | 76.1% | — |

## Math

- Claude 3.7 Sonnet: 35.7 (#137)
- GPT-5.6 Sol: 85.5 (#7)

| Benchmark | Claude 3.7 Sonnet | GPT-5.6 Sol |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 57.8% | 100% |
| LMArena Math | 1336 | 1472 |
| FrontierMath (Tiers 1-3) | — | 89.1% |
| FrontierMath Tier 4 | — | 82.9% |
| ProofBench | — | 83% |
| Omni-MATH | 33% | — |
| LiveBench Math | 79% | — |
| MATH Level 5 | 91.2% | — |
| FrontierMath (Feb 2025 set) | 4.1% | — |
| FrontierMath Erdős | — | 0% |

## Knowledge

- Claude 3.7 Sonnet: 39.7 (#117)
- GPT-5.6 Sol: 64.3 (#16)

| Benchmark | Claude 3.7 Sonnet | GPT-5.6 Sol |
|---|---|---|
| GPQA Diamond | 79.7% | 93.5% |
| LMArena Expert | 1321 | 1517 |
| Humanity's Last Exam | 8% | — |
| SimpleQA Verified | — | 69.7% |
| MMLU-Pro | 78.4% | — |
| Confabulations | 14.7% | — |
| Vectara Hallucination Rate | — | 12.4% |
| GPQA (HELM) | 60.8% | — |

## Multimodal

- Claude 3.7 Sonnet: 33.1 (#90)
- GPT-5.6 Sol: 48.9 (#8)

| Benchmark | Claude 3.7 Sonnet | GPT-5.6 Sol |
|---|---|---|
| LMArena Vision | 1169 | 1281 |
| GeoBench | 68% | — |
| VPCT | 39% | — |
| Blueprint-Bench 2 | — | 33.6% |
| Furniture Assembly | — | 56.7% |
| LMArena Document | — | 1483 |
| SpatialViz-Bench | 33.9% | — |

## Multilingual

- Claude 3.7 Sonnet: 44.3 (#174)
- GPT-5.6 Sol: 55.4 (#32)

| Benchmark | Claude 3.7 Sonnet | GPT-5.6 Sol |
|---|---|---|
| LMArena Non-English | 1296 | 1450 |
| LMArena Chinese | 1299 | 1523 |
| LMArena French | 1302 | 1475 |
| LMArena German | 1301 | 1476 |
| LMArena Japanese | 1268 | 1475 |
| LMArena Korean | 1249 | 1434 |
| LMArena Russian | 1311 | 1466 |
| LMArena Spanish | 1299 | 1449 |

## Instruction Following

- Claude 3.7 Sonnet: 73.1 (#120)
- GPT-5.6 Sol: 77.8 (#16)

| Benchmark | Claude 3.7 Sonnet | GPT-5.6 Sol |
|---|---|---|
| LMArena Instruction Following | 1352 | 1480 |
| LiveBench Instruction Following | 81.3% | — |
| IFEval | 83.4% | — |

## Long Context

- Claude 3.7 Sonnet: 50.2 (#10)
- GPT-5.6 Sol: 45.3 (#42)

| Benchmark | Claude 3.7 Sonnet | GPT-5.6 Sol |
|---|---|---|
| LMArena Longer Query | 1373 | 1479 |
| Fiction.LiveBench | 83.3% | — |

## Writing & Preference

- Claude 3.7 Sonnet: 54.6 (#145)
- GPT-5.6 Sol: 73.5 (#12)

| Benchmark | Claude 3.7 Sonnet | GPT-5.6 Sol |
|---|---|---|
| LMArena Text | 1314 | 1456 |
| LMArena Creative Writing | 1332 | 1450 |
| EQ-Bench Creative Writing | 1412 | 1972 |
| LMArena Multi-Turn | 1339 | 1456 |
| Short-Story Creative Writing | 81.1% | — |
| WildBench | 81.4% | — |
| EQ-Bench 4 | — | 1250 |
| LiveBench Language | 59.9% | — |

## FAQ

### Is Claude 3.7 Sonnet better than GPT-5.6 Sol?

GPT-5.6 Sol is the stronger model overall, scoring 64.4 to 39.4 on the Noometry Index.

### Is Claude 3.7 Sonnet or GPT-5.6 Sol better for coding?

GPT-5.6 Sol scores higher on coding benchmarks: 61.1 versus 41.1 in the Noometry coding category.

### How many benchmarks do Claude 3.7 Sonnet and GPT-5.6 Sol share?

26 benchmarks have published results for both models. Claude 3.7 Sonnet has 58 scored results on Noometry and GPT-5.6 Sol has 64.