Model comparison
Claude 3.7 Sonnet vs GPT-4o
Claude 3.7 Sonnet is the stronger model overall, scoring 39.5 to 28.6 on the Noometry Index.
Last verified . 54 shared benchmarks.
Summary
- They share 54 benchmarks with published results for both. Claude 3.7 Sonnet scores higher in 9 categories and GPT-4o in 1 category; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where Claude 3.7 Sonnet leads 37.5 to 10.6.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 57.8% for Claude 3.7 Sonnet and 6.4% for GPT-4o.
Side by side
| Claude 3.7 Sonnet | GPT-4o | |
|---|---|---|
| Provider | Anthropic | OpenAI |
| Noometry Index | 39.5 | 28.6 |
| Released | 2025-02-24 | 2024-05-13 |
| Weights | Proprietary | Proprietary |
| Context window | — | 128K |
| Max output | — | 16K |
| Input $ / M tokens | — | $2.50 |
| Output $ / M tokens | — | $10 |
| Results tracked | 58 | 72 |
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: 40.6 (#136), GPT-4o: 24.8 (#328)
| Benchmark | Claude 3.7 Sonnet | GPT-4o |
|---|---|---|
| SWE-bench Verified | 61% | 31% |
| SWE-bench Verified (bash only) | 52.8% | 21.6% |
| Aider Polyglot | 64.9% | 45.3% |
| GSO | 3.8% | 0% |
| LiveBench Coding | 74.5% | 51.4% |
| LMArena Coding | 1361 | 1297 |
| CadEval | 54% | 26% |
| WeirdML | — | 25.1% |
| BigCodeBench Instruct | — | 51.1% |
| BigCodeBench Complete | — | 61.1% |
| HumanEval+ | — | 87.2% |
| MBPP+ | — | 72.2% |
Agentic & Tool Use Claude 3.7 Sonnet leads
Claude 3.7 Sonnet: 34.1 (#50), GPT-4o: 21.0 (#141)
| Benchmark | Claude 3.7 Sonnet | GPT-4o |
|---|---|---|
| TheAgentCompany | 30.9% | 8.6% |
| Cybench | 20% | 12.5% |
| METR Time Horizons | 60% | 40.8% |
| GDPval | — | 9.9% |
| DeepResearch Bench | 43.6% | — |
| OSWorld | 35.8% | — |
| BALROG | — | 32.3% |
| LMArena Search | — | 1006 |
Reasoning Claude 3.7 Sonnet leads
Claude 3.7 Sonnet: 18.6 (#277), GPT-4o: 9.4 (#343)
| Benchmark | Claude 3.7 Sonnet | GPT-4o |
|---|---|---|
| ARC-AGI-2 | 0.9% | 0% |
| SimpleBench | 46.4% | 17.8% |
| ARC-AGI-1 | 28.6% | 4.5% |
| EnigmaEval | 4.2% | 0.8% |
| LiveBench Reasoning | 87.8% | 55.8% |
| LMArena Hard Prompts | 1333 | 1281 |
| LiveBench Data Analysis | 74% | 60.9% |
| Epoch Capabilities Index | 141.16 | 128.97 |
| ForecastBench | 61.8 | 57.7 |
| LiveBench | 76.1% | 55.3% |
| CritPt | — | 0% |
| Chess Puzzles | — | 13% |
| DTBench | — | 64.5% |
| LMCA | — | 16.6% |
Math Claude 3.7 Sonnet leads
Claude 3.7 Sonnet: 37.5 (#153), GPT-4o: 10.6 (#312)
| Benchmark | Claude 3.7 Sonnet | GPT-4o |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 57.8% | 6.4% |
| Omni-MATH | 33% | 29.3% |
| LiveBench Math | 79% | 49.5% |
| LMArena Math | 1337 | 1285 |
| MATH Level 5 | 91.2% | 53.3% |
| FrontierMath (Feb 2025 set) | 4.1% | 0.3% |
| FrontierMath (Tiers 1-3) | — | 0.4% |
Knowledge Claude 3.7 Sonnet leads
Claude 3.7 Sonnet: 39.8 (#130), GPT-4o: 28.8 (#242)
| Benchmark | Claude 3.7 Sonnet | GPT-4o |
|---|---|---|
| GPQA Diamond | 79.7% | 49.2% |
| Humanity's Last Exam | 8% | 2.7% |
| MMLU-Pro | 78.4% | 71.3% |
| Confabulations | 14.7% | 15.3% |
| GPQA (HELM) | 60.8% | 52% |
| LMArena Expert | 1321 | 1250 |
| SimpleQA Verified | — | 26% |
| Vectara Hallucination Rate | — | 9.6% |
| MMLU | — | 88.1% |
Multimodal Too close to call
Claude 3.7 Sonnet: 33.7 (#95), GPT-4o: 34.5 (#91)
| Benchmark | Claude 3.7 Sonnet | GPT-4o |
|---|---|---|
| LMArena Vision | 1169 | 1137 |
| GeoBench | 68% | 71% |
| VPCT | 39% | 40% |
| Video-MME | — | 71.9% |
| ScienceQA | — | 88.5% |
| SpatialViz-Bench | 33.9% | — |
Multilingual Too close to call
Claude 3.7 Sonnet: 44.1 (#179), GPT-4o: 43.2 (#186)
| Benchmark | Claude 3.7 Sonnet | GPT-4o |
|---|---|---|
| LMArena Non-English | 1296 | 1283 |
| LMArena Chinese | 1299 | 1277 |
| LMArena French | 1303 | 1304 |
| LMArena German | 1301 | 1282 |
| LMArena Japanese | 1267 | 1257 |
| LMArena Korean | 1249 | 1234 |
| LMArena Russian | 1311 | 1286 |
| LMArena Spanish | 1298 | 1292 |
Instruction Following Claude 3.7 Sonnet leads
Claude 3.7 Sonnet: 72.9 (#125), GPT-4o: 66.6 (#207)
| Benchmark | Claude 3.7 Sonnet | GPT-4o |
|---|---|---|
| LiveBench Instruction Following | 81.3% | 68.6% |
| IFEval | 83.4% | 81.7% |
| LMArena Instruction Following | 1352 | 1278 |
Long Context Claude 3.7 Sonnet leads
Claude 3.7 Sonnet: 50.3 (#10), GPT-4o: 39.4 (#179)
| Benchmark | Claude 3.7 Sonnet | GPT-4o |
|---|---|---|
| Fiction.LiveBench | 83.3% | 66.7% |
| LMArena Longer Query | 1373 | 1289 |
Writing & Preference Claude 3.7 Sonnet leads
Claude 3.7 Sonnet: 54.4 (#150), GPT-4o: 52.6 (#166)
| Benchmark | Claude 3.7 Sonnet | GPT-4o |
|---|---|---|
| LMArena Text | 1314 | 1300 |
| LMArena Creative Writing | 1332 | 1292 |
| Short-Story Creative Writing | 81.1% | 81.8% |
| WildBench | 81.4% | 82.8% |
| LMArena Multi-Turn | 1339 | 1302 |
| LiveBench Language | 59.9% | 47.6% |
| EQ-Bench Creative Writing | 1412 | — |
Frequently asked questions
Is Claude 3.7 Sonnet better than GPT-4o?
Claude 3.7 Sonnet is the stronger model overall, scoring 39.5 to 28.6 on the Noometry Index.
Is Claude 3.7 Sonnet or GPT-4o better for coding?
Claude 3.7 Sonnet scores higher on coding benchmarks: 40.6 versus 24.8 in the Noometry coding category.
How many benchmarks do Claude 3.7 Sonnet and GPT-4o share?
54 benchmarks have published results for both models. Claude 3.7 Sonnet has 58 scored results on Noometry and GPT-4o has 72.