# Claude 3.7 Sonnet vs GPT-4.5

> Claude 3.7 Sonnet is the stronger model overall, scoring 39.4 to 36.9 on the Noometry Index.

- Canonical page: https://noometry.com/compare/claude-3-7-sonnet-vs-gpt-4-5
- Last updated: 2026-10-10
- Shared benchmarks: 41

## Summary

- They share 41 benchmarks with published results for both. Claude 3.7 Sonnet scores higher in 6 categories and GPT-4.5 in 4 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in long context, where Claude 3.7 Sonnet leads 50.2 to 40.4.
- The biggest single-benchmark swing is Aider Polyglot: 64.9% for Claude 3.7 Sonnet and 44.9% for GPT-4.5.

## Snapshot

| | Claude 3.7 Sonnet | GPT-4.5 |
|---|---|---|
| Provider | Anthropic | OpenAI |
| Noometry Index | 39.4 | 36.9 |
| Rank | 154 | 200 |
| Context | — | — |
| Input $/M | — | — |
| Output $/M | — | — |
| Weights | Proprietary | Proprietary |

## Coding

- Claude 3.7 Sonnet: 41.1 (#120)
- GPT-4.5: 41.8 (#102)

| Benchmark | Claude 3.7 Sonnet | GPT-4.5 |
|---|---|---|
| Aider Polyglot | 64.9% | 44.9% |
| LiveBench Coding | 74.5% | 75.2% |
| LMArena Coding | 1361 | 1397 |
| SWE-bench Verified | 61% | — |
| SWE-bench Verified (bash only) | 52.8% | — |
| GSO | 3.8% | — |
| WeirdML | — | 39.4% |
| CadEval | 54% | — |

## Agentic & Tool Use

- Claude 3.7 Sonnet: 33.9 (#46)
- GPT-4.5: 27.6 (#92)

| Benchmark | Claude 3.7 Sonnet | GPT-4.5 |
|---|---|---|
| Cybench | 20% | 17.5% |
| TheAgentCompany | 30.9% | — |
| DeepResearch Bench | 43.6% | — |
| OSWorld | 35.8% | — |
| METR Time Horizons | 60% | — |

## Reasoning

- Claude 3.7 Sonnet: 18.7 (#252)
- GPT-4.5: 14.1 (#320)

| Benchmark | Claude 3.7 Sonnet | GPT-4.5 |
|---|---|---|
| ARC-AGI-2 | 0.9% | 0.8% |
| SimpleBench | 46.4% | 34.5% |
| ARC-AGI-1 | 28.6% | 10.3% |
| EnigmaEval | 4.2% | 3.2% |
| LiveBench Reasoning | 87.8% | 71.1% |
| LMArena Hard Prompts | 1333 | 1404 |
| LiveBench Data Analysis | 74% | 64.3% |
| Epoch Capabilities Index | 141.16 | 136.74 |
| ForecastBench | 61.8 | 61.7 |
| LiveBench | 76.1% | 69% |

## Math

- Claude 3.7 Sonnet: 35.7 (#137)
- GPT-4.5: 31.0 (#200)

| Benchmark | Claude 3.7 Sonnet | GPT-4.5 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 57.8% | 37.8% |
| LiveBench Math | 79% | 69.3% |
| LMArena Math | 1336 | 1412 |
| MATH Level 5 | 91.2% | 78.6% |
| Omni-MATH | 33% | — |
| FrontierMath (Feb 2025 set) | 4.1% | — |

## Knowledge

- Claude 3.7 Sonnet: 39.7 (#117)
- GPT-4.5: 32.4 (#205)

| Benchmark | Claude 3.7 Sonnet | GPT-4.5 |
|---|---|---|
| GPQA Diamond | 79.7% | 68.7% |
| Humanity's Last Exam | 8% | 5.4% |
| Confabulations | 14.7% | 13.6% |
| LMArena Expert | 1321 | 1394 |
| MMLU-Pro | 78.4% | — |
| GPQA (HELM) | 60.8% | — |

## Multimodal

- Claude 3.7 Sonnet: 33.1 (#90)
- GPT-4.5: 37.0 (#68)

| Benchmark | Claude 3.7 Sonnet | GPT-4.5 |
|---|---|---|
| LMArena Vision | 1169 | 1195 |
| VPCT | 39% | 45% |
| GeoBench | 68% | — |
| SpatialViz-Bench | 33.9% | — |

## Multilingual

- Claude 3.7 Sonnet: 44.3 (#174)
- GPT-4.5: 52.7 (#78)

| Benchmark | Claude 3.7 Sonnet | GPT-4.5 |
|---|---|---|
| LMArena Non-English | 1296 | 1413 |
| LMArena Chinese | 1299 | 1421 |
| LMArena French | 1302 | 1417 |
| LMArena German | 1301 | 1457 |
| LMArena Japanese | 1268 | 1416 |
| LMArena Korean | 1249 | 1392 |
| LMArena Russian | 1311 | 1419 |
| LMArena Spanish | 1299 | — |

## Instruction Following

- Claude 3.7 Sonnet: 73.1 (#120)
- GPT-4.5: 72.8 (#129)

| Benchmark | Claude 3.7 Sonnet | GPT-4.5 |
|---|---|---|
| LiveBench Instruction Following | 81.3% | 72.3% |
| LMArena Instruction Following | 1352 | 1404 |
| IFEval | 83.4% | — |

## Long Context

- Claude 3.7 Sonnet: 50.2 (#10)
- GPT-4.5: 40.4 (#150)

| Benchmark | Claude 3.7 Sonnet | GPT-4.5 |
|---|---|---|
| Fiction.LiveBench | 83.3% | 63.9% |
| LMArena Longer Query | 1373 | 1406 |

## Writing & Preference

- Claude 3.7 Sonnet: 54.6 (#145)
- GPT-4.5: 57.1 (#129)

| Benchmark | Claude 3.7 Sonnet | GPT-4.5 |
|---|---|---|
| LMArena Text | 1314 | 1417 |
| LMArena Creative Writing | 1332 | 1394 |
| Short-Story Creative Writing | 81.1% | 75.6% |
| EQ-Bench Creative Writing | 1412 | 1258 |
| LMArena Multi-Turn | 1339 | 1444 |
| LiveBench Language | 59.9% | 61.5% |
| WildBench | 81.4% | — |

## FAQ

### Is Claude 3.7 Sonnet better than GPT-4.5?

Claude 3.7 Sonnet is the stronger model overall, scoring 39.4 to 36.9 on the Noometry Index.

### Is Claude 3.7 Sonnet or GPT-4.5 better for coding?

They score almost the same on coding (41.1 vs 41.8); test both on your own repository before choosing.

### How many benchmarks do Claude 3.7 Sonnet and GPT-4.5 share?

41 benchmarks have published results for both models. Claude 3.7 Sonnet has 58 scored results on Noometry and GPT-4.5 has 42.