Model comparison
Claude 3.7 Sonnet vs GPT-4
Claude 3.7 Sonnet is the stronger model overall, scoring 39.4 to 28.9 on the Noometry Index.
Last verified . 24 shared benchmarks.
Summary
- They share 24 benchmarks with published results for both. Claude 3.7 Sonnet scores higher in 8 categories and GPT-4 in 0 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where Claude 3.7 Sonnet leads 35.7 to 10.1.
- The biggest single-benchmark swing is MATH Level 5: 91.2% for Claude 3.7 Sonnet and 23% for GPT-4.
Side by side
| Claude 3.7 Sonnet | GPT-4 | |
|---|---|---|
| Provider | Anthropic | OpenAI |
| Noometry Index | 39.4 | 28.9 |
| Released | 2025-02-24 | 2023-03-14 |
| Weights | Proprietary | Proprietary |
| Context window | — | 8K |
| Max output | — | 8K |
| Input $ / M tokens | — | $30 |
| Output $ / M tokens | — | $60 |
| Results tracked | 58 | 38 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Claude 3.7 Sonnet leads
Claude 3.7 Sonnet: 41.1 (#120), GPT-4: 31.2 (#277)
| Benchmark | Claude 3.7 Sonnet | GPT-4 |
|---|---|---|
| LMArena Coding | 1361 | 1254 |
| SWE-bench Verified | 61% | — |
| SWE-bench Verified (bash only) | 52.8% | — |
| Aider Polyglot | 64.9% | — |
| GSO | 3.8% | — |
| WeirdML | — | 12.4% |
| BigCodeBench Instruct | — | 46% |
| LiveBench Coding | 74.5% | — |
| BigCodeBench Complete | — | 57.2% |
| CadEval | 54% | — |
| HumanEval+ | — | 79.3% |
Agentic & Tool Use Not comparable
Claude 3.7 Sonnet: 33.9 (#46), GPT-4: —
| Benchmark | Claude 3.7 Sonnet | GPT-4 |
|---|---|---|
| METR Time Horizons | 60% | 36.1% |
| TheAgentCompany | 30.9% | — |
| Cybench | 20% | — |
| DeepResearch Bench | 43.6% | — |
| OSWorld | 35.8% | — |
Reasoning Claude 3.7 Sonnet leads
Claude 3.7 Sonnet: 18.7 (#252), GPT-4: 17.3 (#285)
| Benchmark | Claude 3.7 Sonnet | GPT-4 |
|---|---|---|
| LMArena Hard Prompts | 1333 | 1241 |
| Epoch Capabilities Index | 141.16 | 125.88 |
| ForecastBench | 61.8 | 57.8 |
| ARC-AGI-2 | 0.9% | — |
| SimpleBench | 46.4% | — |
| ARC-AGI-1 | 28.6% | — |
| Chess Puzzles | — | 4% |
| EnigmaEval | 4.2% | — |
| LiveBench Reasoning | 87.8% | — |
| Mystery Game Puzzles | — | 12% |
| DTBench | — | 62.7% |
| LiveBench Data Analysis | 74% | — |
| LMCA | — | 17.1% |
| BIG-Bench Hard | — | 75.1% |
| HellaSwag | — | 95.3% |
| LiveBench | 76.1% | — |
| WinoGrande | — | 87.5% |
Math Claude 3.7 Sonnet leads
Claude 3.7 Sonnet: 35.7 (#137), GPT-4: 10.1 (#304)
| Benchmark | Claude 3.7 Sonnet | GPT-4 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 57.8% | 1.1% |
| LMArena Math | 1336 | 1269 |
| MATH Level 5 | 91.2% | 23% |
| Omni-MATH | 33% | — |
| LiveBench Math | 79% | — |
| FrontierMath (Feb 2025 set) | 4.1% | — |
| GSM8K | — | 92% |
Knowledge Claude 3.7 Sonnet leads
Claude 3.7 Sonnet: 39.7 (#117), GPT-4: 18.4 (#277)
| Benchmark | Claude 3.7 Sonnet | GPT-4 |
|---|---|---|
| GPQA Diamond | 79.7% | 35.7% |
| LMArena Expert | 1321 | 1211 |
| Humanity's Last Exam | 8% | — |
| MMLU-Pro | 78.4% | — |
| Confabulations | 14.7% | — |
| GPQA (HELM) | 60.8% | — |
| MMLU | — | 86.4% |
| TriviaQA | — | 84.8% |
Multimodal Not comparable
Claude 3.7 Sonnet: 33.1 (#90), GPT-4: —
| Benchmark | Claude 3.7 Sonnet | GPT-4 |
|---|---|---|
| LMArena Vision | 1169 | — |
| GeoBench | 68% | — |
| VPCT | 39% | — |
| SpatialViz-Bench | 33.9% | — |
Multilingual Claude 3.7 Sonnet leads
Claude 3.7 Sonnet: 44.3 (#174), GPT-4: 40.8 (#210)
| Benchmark | Claude 3.7 Sonnet | GPT-4 |
|---|---|---|
| LMArena Non-English | 1296 | 1246 |
| LMArena Chinese | 1299 | 1242 |
| LMArena French | 1302 | 1282 |
| LMArena German | 1301 | 1251 |
| LMArena Japanese | 1268 | 1209 |
| LMArena Korean | 1249 | 1184 |
| LMArena Russian | 1311 | 1251 |
| LMArena Spanish | 1299 | 1261 |
Instruction Following Claude 3.7 Sonnet leads
Claude 3.7 Sonnet: 73.1 (#120), GPT-4: 65.5 (#217)
| Benchmark | Claude 3.7 Sonnet | GPT-4 |
|---|---|---|
| LMArena Instruction Following | 1352 | 1241 |
| LiveBench Instruction Following | 81.3% | — |
| IFEval | 83.4% | — |
Long Context Claude 3.7 Sonnet leads
Claude 3.7 Sonnet: 50.2 (#10), GPT-4: 37.7 (#207)
| Benchmark | Claude 3.7 Sonnet | GPT-4 |
|---|---|---|
| LMArena Longer Query | 1373 | 1244 |
| Fiction.LiveBench | 83.3% | — |
Writing & Preference Claude 3.7 Sonnet leads
Claude 3.7 Sonnet: 54.6 (#145), GPT-4: 35.0 (#263)
| Benchmark | Claude 3.7 Sonnet | GPT-4 |
|---|---|---|
| LMArena Text | 1314 | 1263 |
| LMArena Creative Writing | 1332 | 1244 |
| EQ-Bench Creative Writing | 1412 | 752 |
| LMArena Multi-Turn | 1339 | 1257 |
| Short-Story Creative Writing | 81.1% | — |
| WildBench | 81.4% | — |
| LiveBench Language | 59.9% | — |
Frequently asked questions
Is Claude 3.7 Sonnet better than GPT-4?
Claude 3.7 Sonnet is the stronger model overall, scoring 39.4 to 28.9 on the Noometry Index.
Is Claude 3.7 Sonnet or GPT-4 better for coding?
Claude 3.7 Sonnet scores higher on coding benchmarks: 41.1 versus 31.2 in the Noometry coding category.
How many benchmarks do Claude 3.7 Sonnet and GPT-4 share?
24 benchmarks have published results for both models. Claude 3.7 Sonnet has 58 scored results on Noometry and GPT-4 has 38.