# Claude 3.7 Sonnet vs GPT-4o mini

> Claude 3.7 Sonnet is the stronger model overall, scoring 39.5 to 25.5 on the Noometry Index.

- Canonical page: https://noometry.com/compare/claude-3-7-sonnet-vs-gpt-4o-mini
- Last updated: 2026-10-12
- Shared benchmarks: 42

## Summary

- They share 42 benchmarks with published results for both. Claude 3.7 Sonnet scores higher in 10 categories and GPT-4o mini in 0 categories; 10 gaps are clear of the uncertainty.
- The widest gap is in math, where Claude 3.7 Sonnet leads 37.5 to 10.4.
- The biggest single-benchmark swing is Aider Polyglot: 64.9% for Claude 3.7 Sonnet and 3.6% for GPT-4o mini.

## Snapshot

| | Claude 3.7 Sonnet | GPT-4o mini |
|---|---|---|
| Provider | Anthropic | OpenAI |
| Noometry Index | 39.5 | 25.5 |
| Rank | 164 | 343 |
| Context | — | 128K |
| Input $/M | — | $0.15 |
| Output $/M | — | $0.60 |
| Weights | Proprietary | Proprietary |

## Coding

- Claude 3.7 Sonnet: 40.6 (#136)
- GPT-4o mini: 22.0 (#335)

| Benchmark | Claude 3.7 Sonnet | GPT-4o mini |
|---|---|---|
| Aider Polyglot | 64.9% | 3.6% |
| LiveBench Coding | 74.5% | 43.1% |
| LMArena Coding | 1361 | 1290 |
| SWE-bench Verified | 61% | — |
| SWE-bench Verified (bash only) | 52.8% | — |
| GSO | 3.8% | — |
| WeirdML | — | 11.8% |
| BigCodeBench Instruct | — | 46.1% |
| BigCodeBench Complete | — | 57.4% |
| CadEval | 54% | — |
| HumanEval+ | — | 83.5% |
| MBPP+ | — | 72.2% |

## Agentic & Tool Use

- Claude 3.7 Sonnet: 34.1 (#50)
- GPT-4o mini: 27.5 (#101)

| Benchmark | Claude 3.7 Sonnet | GPT-4o mini |
|---|---|---|
| TheAgentCompany | 30.9% | — |
| Cybench | 20% | — |
| DeepResearch Bench | 43.6% | — |
| OSWorld | 35.8% | — |
| BALROG | — | 17.4% |
| METR Time Horizons | 60% | — |

## Reasoning

- Claude 3.7 Sonnet: 18.6 (#277)
- GPT-4o mini: 8.7 (#347)

| Benchmark | Claude 3.7 Sonnet | GPT-4o mini |
|---|---|---|
| ARC-AGI-2 | 0.9% | 0% |
| SimpleBench | 46.4% | 10.7% |
| LiveBench Reasoning | 87.8% | 32.8% |
| LMArena Hard Prompts | 1333 | 1267 |
| LiveBench Data Analysis | 74% | 50% |
| Epoch Capabilities Index | 141.16 | 126.56 |
| LiveBench | 76.1% | 41.3% |
| Kagi LLM Benchmark | — | 28.8% |
| ARC-AGI-1 | 28.6% | — |
| Chess Puzzles | — | 0% |
| EnigmaEval | 4.2% | — |
| Mystery Game Puzzles | — | 12% |
| DTBench | — | 54.4% |
| LMCA | — | 10.4% |
| ForecastBench | 61.8 | — |
| PIQA | — | 88.7% |

## Math

- Claude 3.7 Sonnet: 37.5 (#153)
- GPT-4o mini: 10.4 (#314)

| Benchmark | Claude 3.7 Sonnet | GPT-4o mini |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 57.8% | 6.9% |
| Omni-MATH | 33% | 28% |
| LiveBench Math | 79% | 36.3% |
| LMArena Math | 1337 | 1267 |
| MATH Level 5 | 91.2% | 52.6% |
| FrontierMath (Tiers 1-3) | — | 0.7% |
| FrontierMath (Feb 2025 set) | 4.1% | — |
| GSM8K | — | 91.3% |

## Knowledge

- Claude 3.7 Sonnet: 39.8 (#130)
- GPT-4o mini: 17.7 (#284)

| Benchmark | Claude 3.7 Sonnet | GPT-4o mini |
|---|---|---|
| GPQA Diamond | 79.7% | 37.7% |
| MMLU-Pro | 78.4% | 60.3% |
| Confabulations | 14.7% | 37.2% |
| GPQA (HELM) | 60.8% | 36.8% |
| LMArena Expert | 1321 | 1235 |
| Humanity's Last Exam | 8% | — |
| SimpleQA Verified | — | 8.3% |
| BoolQ | — | 88.7% |
| MMLU | — | 81.8% |

## Multimodal

- Claude 3.7 Sonnet: 33.7 (#95)
- GPT-4o mini: 25.9 (#122)

| Benchmark | Claude 3.7 Sonnet | GPT-4o mini |
|---|---|---|
| LMArena Vision | 1169 | 1066 |
| GeoBench | 68% | 64% |
| VPCT | 39% | 34% |
| Video-MME | — | 64.8% |
| SpatialViz-Bench | 33.9% | — |

## Multilingual

- Claude 3.7 Sonnet: 44.1 (#179)
- GPT-4o mini: 42.0 (#199)

| Benchmark | Claude 3.7 Sonnet | GPT-4o mini |
|---|---|---|
| LMArena Non-English | 1296 | 1266 |
| LMArena Chinese | 1299 | 1265 |
| LMArena French | 1303 | 1297 |
| LMArena German | 1301 | 1272 |
| LMArena Japanese | 1267 | 1216 |
| LMArena Korean | 1249 | 1195 |
| LMArena Russian | 1311 | 1275 |
| LMArena Spanish | 1298 | 1276 |

## Instruction Following

- Claude 3.7 Sonnet: 72.9 (#125)
- GPT-4o mini: 61.9 (#239)

| Benchmark | Claude 3.7 Sonnet | GPT-4o mini |
|---|---|---|
| LiveBench Instruction Following | 81.3% | 56.8% |
| IFEval | 83.4% | 78.2% |
| LMArena Instruction Following | 1352 | 1258 |

## Long Context

- Claude 3.7 Sonnet: 50.3 (#10)
- GPT-4o mini: 39.1 (#186)

| Benchmark | Claude 3.7 Sonnet | GPT-4o mini |
|---|---|---|
| LMArena Longer Query | 1373 | 1289 |
| Fiction.LiveBench | 83.3% | — |

## Writing & Preference

- Claude 3.7 Sonnet: 54.4 (#150)
- GPT-4o mini: 39.5 (#248)

| Benchmark | Claude 3.7 Sonnet | GPT-4o mini |
|---|---|---|
| LMArena Text | 1314 | 1286 |
| LMArena Creative Writing | 1332 | 1268 |
| Short-Story Creative Writing | 81.1% | 67.2% |
| EQ-Bench Creative Writing | 1412 | 873 |
| WildBench | 81.4% | 79.1% |
| LMArena Multi-Turn | 1339 | 1285 |
| LiveBench Language | 59.9% | 28.6% |

## FAQ

### Is Claude 3.7 Sonnet better than GPT-4o mini?

Claude 3.7 Sonnet is the stronger model overall, scoring 39.5 to 25.5 on the Noometry Index.

### Is Claude 3.7 Sonnet or GPT-4o mini better for coding?

Claude 3.7 Sonnet scores higher on coding benchmarks: 40.6 versus 22.0 in the Noometry coding category.

### How many benchmarks do Claude 3.7 Sonnet and GPT-4o mini share?

42 benchmarks have published results for both models. Claude 3.7 Sonnet has 58 scored results on Noometry and GPT-4o mini has 60.