Model comparison
Claude 3.7 Sonnet vs Claude Sonnet 4
Claude Sonnet 4 is the stronger model overall, scoring 40.8 to 39.5 on the Noometry Index.
Last verified . 48 shared benchmarks.
Summary
- They share 48 benchmarks with published results for both. Claude 3.7 Sonnet scores higher in 3 categories and Claude Sonnet 4 in 7 categories; 10 gaps are clear of the uncertainty.
- The widest gap is in long context, where Claude 3.7 Sonnet leads 50.3 to 33.7.
- The biggest single-benchmark swing is Fiction.LiveBench: 83.3% for Claude 3.7 Sonnet and 46.9% for Claude Sonnet 4.
Side by side
| Claude 3.7 Sonnet | Claude Sonnet 4 | |
|---|---|---|
| Provider | Anthropic | Anthropic |
| Noometry Index | 39.5 | 40.8 |
| Released | 2025-02-24 | 2025-05-22 |
| Weights | Proprietary | Proprietary |
| Context window | — | 200K |
| Max output | — | 64K |
| Input $ / M tokens | — | $3 |
| Output $ / M tokens | — | $15 |
| Results tracked | 58 | 58 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Claude Sonnet 4 leads
Claude 3.7 Sonnet: 40.6 (#136), Claude Sonnet 4: 43.5 (#88)
| Benchmark | Claude 3.7 Sonnet | Claude Sonnet 4 |
|---|---|---|
| SWE-bench Verified (bash only) | 52.8% | 64.9% |
| Aider Polyglot | 64.9% | 61.3% |
| GSO | 3.8% | 4.9% |
| LMArena Coding | 1361 | 1414 |
| SWE-bench Verified | 61% | — |
| SciCode | — | 40% |
| WeirdML | — | 46.1% |
| LiveBench Coding | 74.5% | — |
| CadEval | 54% | — |
| ALE-Bench | — | 655.35 |
Agentic & Tool Use Claude Sonnet 4 leads
Claude 3.7 Sonnet: 34.1 (#50), Claude Sonnet 4: 38.5 (#31)
| Benchmark | Claude 3.7 Sonnet | Claude Sonnet 4 |
|---|---|---|
| TheAgentCompany | 30.9% | 33.1% |
| Cybench | 20% | 35% |
| DeepResearch Bench | 43.6% | 46.6% |
| OSWorld | 35.8% | 43.9% |
| METR Time Horizons | 60% | 62% |
Reasoning Claude Sonnet 4 leads
Claude 3.7 Sonnet: 18.6 (#277), Claude Sonnet 4: 22.9 (#187)
| Benchmark | Claude 3.7 Sonnet | Claude Sonnet 4 |
|---|---|---|
| ARC-AGI-2 | 0.9% | 5.9% |
| SimpleBench | 46.4% | 45.5% |
| ARC-AGI-1 | 28.6% | 40% |
| EnigmaEval | 4.2% | 3.1% |
| LMArena Hard Prompts | 1333 | 1372 |
| Epoch Capabilities Index | 141.16 | 141.69 |
| ForecastBench | 61.8 | 60.2 |
| Kagi LLM Benchmark | — | 73% |
| CritPt | — | 0.3% |
| LiveBench Reasoning | 87.8% | — |
| DTBench | — | 77.1% |
| LiveBench Data Analysis | 74% | — |
| LMCA | — | 29% |
| LiveBench | 76.1% | — |
Math Claude Sonnet 4 leads
Claude 3.7 Sonnet: 37.5 (#153), Claude Sonnet 4: 43.3 (#80)
| Benchmark | Claude 3.7 Sonnet | Claude Sonnet 4 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 57.8% | 71.1% |
| Omni-MATH | 33% | 60.2% |
| LMArena Math | 1337 | 1375 |
| MATH Level 5 | 91.2% | 84.4% |
| FrontierMath (Feb 2025 set) | 4.1% | 4.1% |
| LiveBench Math | 79% | — |
| FrontierMath Tier 4 (v1) | — | 0% |
Knowledge Claude Sonnet 4 leads
Claude 3.7 Sonnet: 39.8 (#130), Claude Sonnet 4: 41.8 (#108)
| Benchmark | Claude 3.7 Sonnet | Claude Sonnet 4 |
|---|---|---|
| GPQA Diamond | 79.7% | 79.2% |
| Humanity's Last Exam | 8% | 7.8% |
| MMLU-Pro | 78.4% | 84.3% |
| Confabulations | 14.7% | 13.2% |
| GPQA (HELM) | 60.8% | 70.6% |
| LMArena Expert | 1321 | 1372 |
| Vectara Hallucination Rate | — | 10.3% |
Multimodal Claude 3.7 Sonnet leads
Claude 3.7 Sonnet: 33.7 (#95), Claude Sonnet 4: 26.2 (#121)
| Benchmark | Claude 3.7 Sonnet | Claude Sonnet 4 |
|---|---|---|
| LMArena Vision | 1169 | 1191 |
| GeoBench | 68% | 37% |
| VPCT | 39% | 34% |
| MindCube | — | 44.8% |
| SpatialViz-Bench | 33.9% | — |
Multilingual Claude Sonnet 4 leads
Claude 3.7 Sonnet: 44.1 (#179), Claude Sonnet 4: 46.7 (#156)
| Benchmark | Claude 3.7 Sonnet | Claude Sonnet 4 |
|---|---|---|
| LMArena Non-English | 1296 | 1333 |
| LMArena Chinese | 1299 | 1350 |
| LMArena French | 1303 | 1363 |
| LMArena German | 1301 | 1331 |
| LMArena Japanese | 1267 | 1302 |
| LMArena Korean | 1249 | 1291 |
| LMArena Russian | 1311 | 1355 |
| LMArena Spanish | 1298 | 1357 |
Instruction Following Claude 3.7 Sonnet leads
Claude 3.7 Sonnet: 72.9 (#125), Claude Sonnet 4: 71.7 (#145)
| Benchmark | Claude 3.7 Sonnet | Claude Sonnet 4 |
|---|---|---|
| IFEval | 83.4% | 84% |
| LMArena Instruction Following | 1352 | 1376 |
| LiveBench Instruction Following | 81.3% | — |
Long Context Claude 3.7 Sonnet leads
Claude 3.7 Sonnet: 50.3 (#10), Claude Sonnet 4: 33.7 (#259)
| Benchmark | Claude 3.7 Sonnet | Claude Sonnet 4 |
|---|---|---|
| Fiction.LiveBench | 83.3% | 46.9% |
| LMArena Longer Query | 1373 | 1398 |
Writing & Preference Claude Sonnet 4 leads
Claude 3.7 Sonnet: 54.4 (#150), Claude Sonnet 4: 57.1 (#132)
| Benchmark | Claude 3.7 Sonnet | Claude Sonnet 4 |
|---|---|---|
| LMArena Text | 1314 | 1351 |
| LMArena Creative Writing | 1332 | 1345 |
| Short-Story Creative Writing | 81.1% | 81.4% |
| EQ-Bench Creative Writing | 1412 | 1483 |
| WildBench | 81.4% | 83.8% |
| LMArena Multi-Turn | 1339 | 1376 |
| LiveBench Language | 59.9% | — |
Frequently asked questions
Is Claude 3.7 Sonnet better than Claude Sonnet 4?
Claude Sonnet 4 is the stronger model overall, scoring 40.8 to 39.5 on the Noometry Index.
Is Claude 3.7 Sonnet or Claude Sonnet 4 better for coding?
Claude Sonnet 4 scores higher on coding benchmarks: 43.5 versus 40.6 in the Noometry coding category.
How many benchmarks do Claude 3.7 Sonnet and Claude Sonnet 4 share?
48 benchmarks have published results for both models. Claude 3.7 Sonnet has 58 scored results on Noometry and Claude Sonnet 4 has 58.