Model comparison
Claude 3.7 Sonnet vs GPT-5.1
GPT-5.1 is the stronger model overall, scoring 49.0 to 39.5 on the Noometry Index.
Last verified . 45 shared benchmarks.
Summary
- They share 45 benchmarks with published results for both. Claude 3.7 Sonnet scores higher in 2 categories and GPT-5.1 in 8 categories; 10 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where GPT-5.1 leads 39.8 to 18.6.
- The biggest single-benchmark swing is ARC-AGI-1: 28.6% for Claude 3.7 Sonnet and 72.8% for GPT-5.1.
Side by side
| Claude 3.7 Sonnet | GPT-5.1 | |
|---|---|---|
| Provider | Anthropic | OpenAI |
| Noometry Index | 39.5 | 49.0 |
| Released | 2025-02-24 | 2025-11-13 |
| Weights | Proprietary | Proprietary |
| Context window | — | 400K |
| Max output | — | 128K |
| Input $ / M tokens | — | $1.25 |
| Output $ / M tokens | — | $10 |
| Results tracked | 58 | 63 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-5.1 leads
Claude 3.7 Sonnet: 40.6 (#136), GPT-5.1: 46.4 (#66)
| Benchmark | Claude 3.7 Sonnet | GPT-5.1 |
|---|---|---|
| SWE-bench Verified | 61% | 68% |
| SWE-bench Verified (bash only) | 52.8% | 66% |
| GSO | 3.8% | 13.7% |
| LiveBench Coding | 74.5% | 72.5% |
| LMArena Coding | 1361 | 1454 |
| Aider Polyglot | 64.9% | — |
| LMArena WebDev | — | 1395 |
| SciCode | — | 43.3% |
| WeirdML | — | 60.8% |
| CadEval | 54% | — |
| ALE-Bench | — | 1,192 |
Agentic & Tool Use Claude 3.7 Sonnet leads
Claude 3.7 Sonnet: 34.1 (#50), GPT-5.1: 32.7 (#60)
| Benchmark | Claude 3.7 Sonnet | GPT-5.1 |
|---|---|---|
| DeepResearch Bench | 43.6% | 42.8% |
| Terminal-Bench | — | 47.6% |
| TheAgentCompany | 30.9% | — |
| Cybench | 20% | — |
| OSWorld | 35.8% | — |
| LMArena Search | — | 1199 |
| METR Time Horizons | 60% | — |
| Vending-Bench 2 | — | 1,473 |
Reasoning GPT-5.1 leads
Claude 3.7 Sonnet: 18.6 (#277), GPT-5.1: 39.8 (#58)
| Benchmark | Claude 3.7 Sonnet | GPT-5.1 |
|---|---|---|
| ARC-AGI-2 | 0.9% | 17.6% |
| SimpleBench | 46.4% | 53.2% |
| ARC-AGI-1 | 28.6% | 72.8% |
| EnigmaEval | 4.2% | 11.2% |
| LiveBench Reasoning | 87.8% | 95.8% |
| LMArena Hard Prompts | 1333 | 1457 |
| LiveBench Data Analysis | 74% | 72.1% |
| Epoch Capabilities Index | 141.16 | 149.64 |
| ForecastBench | 61.8 | 58.1 |
| LiveBench | 76.1% | 78.8% |
| CritPt | — | 4.9% |
| Chess Puzzles | — | 32% |
| Mystery Game Puzzles | — | 19% |
| DTBench | — | 90.1% |
| LMCA | — | 43.9% |
Math GPT-5.1 leads
Claude 3.7 Sonnet: 37.5 (#153), GPT-5.1: 52.2 (#51)
| Benchmark | Claude 3.7 Sonnet | GPT-5.1 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 57.8% | 88.6% |
| Omni-MATH | 33% | 46.4% |
| LiveBench Math | 79% | 94.5% |
| LMArena Math | 1337 | 1447 |
| FrontierMath (Feb 2025 set) | 4.1% | 31% |
| MATH Level 5 | 91.2% | — |
| FrontierMath Tier 4 (v1) | — | 12.5% |
Knowledge GPT-5.1 leads
Claude 3.7 Sonnet: 39.8 (#130), GPT-5.1: 50.6 (#71)
| Benchmark | Claude 3.7 Sonnet | GPT-5.1 |
|---|---|---|
| GPQA Diamond | 79.7% | 87.6% |
| Humanity's Last Exam | 8% | 23.7% |
| MMLU-Pro | 78.4% | 57.9% |
| GPQA (HELM) | 60.8% | 44.2% |
| LMArena Expert | 1321 | 1470 |
| SimpleQA Verified | — | 48% |
| Confabulations | 14.7% | — |
| Vectara Hallucination Rate | — | 10.9% |
Multimodal GPT-5.1 leads
Claude 3.7 Sonnet: 33.7 (#95), GPT-5.1: 44.8 (#19)
| Benchmark | Claude 3.7 Sonnet | GPT-5.1 |
|---|---|---|
| LMArena Vision | 1169 | 1250 |
| VPCT | 39% | 58.7% |
| GeoBench | 68% | — |
| LMArena Document | — | 1403 |
| SpatialViz-Bench | 33.9% | — |
Multilingual GPT-5.1 leads
Claude 3.7 Sonnet: 44.1 (#179), GPT-5.1: 53.8 (#56)
| Benchmark | Claude 3.7 Sonnet | GPT-5.1 |
|---|---|---|
| LMArena Non-English | 1296 | 1431 |
| LMArena Chinese | 1299 | 1495 |
| LMArena French | 1303 | 1450 |
| LMArena German | 1301 | 1438 |
| LMArena Japanese | 1267 | 1453 |
| LMArena Korean | 1249 | 1401 |
| LMArena Russian | 1311 | 1435 |
| LMArena Spanish | 1298 | 1433 |
Instruction Following GPT-5.1 leads
Claude 3.7 Sonnet: 72.9 (#125), GPT-5.1: 83.9 (#1)
| Benchmark | Claude 3.7 Sonnet | GPT-5.1 |
|---|---|---|
| LiveBench Instruction Following | 81.3% | 93.3% |
| IFEval | 83.4% | 93.5% |
| LMArena Instruction Following | 1352 | 1443 |
Long Context Claude 3.7 Sonnet leads
Claude 3.7 Sonnet: 50.3 (#10), GPT-5.1: 47.6 (#14)
| Benchmark | Claude 3.7 Sonnet | GPT-5.1 |
|---|---|---|
| LMArena Longer Query | 1373 | 1447 |
| Fiction.LiveBench | 83.3% | — |
| CL-bench | — | 23.7% |
| CL-bench Life | — | 17.3% |
Writing & Preference GPT-5.1 leads
Claude 3.7 Sonnet: 54.4 (#150), GPT-5.1: 64.5 (#55)
| Benchmark | Claude 3.7 Sonnet | GPT-5.1 |
|---|---|---|
| LMArena Text | 1314 | 1443 |
| LMArena Creative Writing | 1332 | 1427 |
| WildBench | 81.4% | 86.3% |
| LMArena Multi-Turn | 1339 | 1450 |
| LiveBench Language | 59.9% | 80.2% |
| Short-Story Creative Writing | 81.1% | — |
| EQ-Bench Creative Writing | 1412 | — |
Frequently asked questions
Is Claude 3.7 Sonnet better than GPT-5.1?
GPT-5.1 is the stronger model overall, scoring 49.0 to 39.5 on the Noometry Index.
Is Claude 3.7 Sonnet or GPT-5.1 better for coding?
GPT-5.1 scores higher on coding benchmarks: 46.4 versus 40.6 in the Noometry coding category.
How many benchmarks do Claude 3.7 Sonnet and GPT-5.1 share?
45 benchmarks have published results for both models. Claude 3.7 Sonnet has 58 scored results on Noometry and GPT-5.1 has 63.