Model comparison
Claude 3.7 Sonnet vs GPT-5 Mini
GPT-5 Mini is the stronger model overall, scoring 41.7 to 39.4 on the Noometry Index.
Last verified . 40 shared benchmarks.
Summary
- They share 40 benchmarks with published results for both. Claude 3.7 Sonnet scores higher in 3 categories and GPT-5 Mini in 7 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in math, where GPT-5 Mini leads 46.3 to 35.7.
- The biggest single-benchmark swing is Omni-MATH: 33% for Claude 3.7 Sonnet and 72.2% for GPT-5 Mini.
Side by side
| Claude 3.7 Sonnet | GPT-5 Mini | |
|---|---|---|
| Provider | Anthropic | OpenAI |
| Noometry Index | 39.4 | 41.7 |
| Released | 2025-02-24 | 2025-08-07 |
| Weights | Proprietary | Proprietary |
| Context window | — | 400K |
| Max output | — | 128K |
| Input $ / M tokens | — | $0.25 |
| Output $ / M tokens | — | $2 |
| Results tracked | 58 | 60 |
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-5 Mini: 40.0 (#134)
| Benchmark | Claude 3.7 Sonnet | GPT-5 Mini |
|---|---|---|
| SWE-bench Verified | 61% | 64.7% |
| SWE-bench Verified (bash only) | 52.8% | 59.8% |
| LMArena Coding | 1361 | 1406 |
| Aider Polyglot | 64.9% | — |
| SWE-bench Multilingual | — | 39.7% |
| SciCode | — | 39.2% |
| GSO | 3.8% | — |
| WeirdML | — | 52.7% |
| LiveBench Coding | 74.5% | — |
| CadEval | 54% | — |
| ALE-Bench | — | 799.77 |
| AlgoTune | — | 1.38 |
Agentic & Tool Use Claude 3.7 Sonnet leads
Claude 3.7 Sonnet: 33.9 (#46), GPT-5 Mini: 31.0 (#64)
| Benchmark | Claude 3.7 Sonnet | GPT-5 Mini |
|---|---|---|
| Terminal-Bench | — | 34.8% |
| Berkeley Function Calling Leaderboard | — | 55.5% |
| TheAgentCompany | 30.9% | — |
| Cybench | 20% | — |
| DeepResearch Bench | 43.6% | — |
| OSWorld | 35.8% | — |
| METR Time Horizons | 60% | — |
| Vending-Bench 2 | — | -31.18 |
Reasoning GPT-5 Mini leads
Claude 3.7 Sonnet: 18.7 (#252), GPT-5 Mini: 24.1 (#153)
| Benchmark | Claude 3.7 Sonnet | GPT-5 Mini |
|---|---|---|
| ARC-AGI-2 | 0.9% | 4.4% |
| ARC-AGI-1 | 28.6% | 54.3% |
| EnigmaEval | 4.2% | 8.2% |
| LMArena Hard Prompts | 1333 | 1380 |
| Epoch Capabilities Index | 141.16 | 145.51 |
| ForecastBench | 61.8 | 61 |
| SimpleBench | 46.4% | — |
| Kagi LLM Benchmark | — | 70.3% |
| CritPt | — | 0% |
| Chess Puzzles | — | 30% |
| LiveBench Reasoning | 87.8% | — |
| Mystery Game Puzzles | — | 10% |
| DTBench | — | 80.5% |
| LiveBench Data Analysis | 74% | — |
| LMCA | — | 34.2% |
| LiveBench | 76.1% | — |
Math GPT-5 Mini leads
Claude 3.7 Sonnet: 35.7 (#137), GPT-5 Mini: 46.3 (#59)
| Benchmark | Claude 3.7 Sonnet | GPT-5 Mini |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 57.8% | 86.7% |
| Omni-MATH | 33% | 72.2% |
| LMArena Math | 1336 | 1377 |
| MATH Level 5 | 91.2% | 97.8% |
| FrontierMath (Feb 2025 set) | 4.1% | 27.2% |
| FrontierMath (Tiers 1-3) | — | 46.7% |
| FrontierMath Tier 4 | — | 12.2% |
| ProofBench | — | 9% |
| LiveBench Math | 79% | — |
| FrontierMath Tier 4 (v1) | — | 6.3% |
Knowledge GPT-5 Mini leads
Claude 3.7 Sonnet: 39.7 (#117), GPT-5 Mini: 45.5 (#80)
| Benchmark | Claude 3.7 Sonnet | GPT-5 Mini |
|---|---|---|
| GPQA Diamond | 79.7% | 75% |
| Humanity's Last Exam | 8% | 19.4% |
| MMLU-Pro | 78.4% | 83.5% |
| Confabulations | 14.7% | 13.3% |
| GPQA (HELM) | 60.8% | 75.6% |
| LMArena Expert | 1321 | 1377 |
| SimpleQA Verified | — | 21.6% |
| Vectara Hallucination Rate | — | 12.9% |
Multimodal GPT-5 Mini leads
Claude 3.7 Sonnet: 33.1 (#90), GPT-5 Mini: 35.0 (#80)
| Benchmark | Claude 3.7 Sonnet | GPT-5 Mini |
|---|---|---|
| LMArena Vision | 1169 | 1202 |
| VPCT | 39% | 40.2% |
| GeoBench | 68% | — |
| SpatialViz-Bench | 33.9% | — |
Multilingual GPT-5 Mini leads
Claude 3.7 Sonnet: 44.3 (#174), GPT-5 Mini: 49.1 (#131)
| Benchmark | Claude 3.7 Sonnet | GPT-5 Mini |
|---|---|---|
| LMArena Non-English | 1296 | 1363 |
| LMArena Chinese | 1299 | 1387 |
| LMArena French | 1302 | 1385 |
| LMArena German | 1301 | 1366 |
| LMArena Japanese | 1268 | 1342 |
| LMArena Korean | 1249 | 1306 |
| LMArena Russian | 1311 | 1362 |
| LMArena Spanish | 1299 | 1355 |
Instruction Following GPT-5 Mini leads
Claude 3.7 Sonnet: 73.1 (#120), GPT-5 Mini: 76.4 (#43)
| Benchmark | Claude 3.7 Sonnet | GPT-5 Mini |
|---|---|---|
| IFEval | 83.4% | 92.7% |
| LMArena Instruction Following | 1352 | 1357 |
| LiveBench Instruction Following | 81.3% | — |
Long Context Claude 3.7 Sonnet leads
Claude 3.7 Sonnet: 50.2 (#10), GPT-5 Mini: 41.9 (#127)
| Benchmark | Claude 3.7 Sonnet | GPT-5 Mini |
|---|---|---|
| Fiction.LiveBench | 83.3% | 69.4% |
| LMArena Longer Query | 1373 | 1355 |
Writing & Preference Too close to call
Claude 3.7 Sonnet: 54.6 (#145), GPT-5 Mini: 55.4 (#143)
| Benchmark | Claude 3.7 Sonnet | GPT-5 Mini |
|---|---|---|
| LMArena Text | 1314 | 1373 |
| LMArena Creative Writing | 1332 | 1325 |
| Short-Story Creative Writing | 81.1% | 83.1% |
| EQ-Bench Creative Writing | 1412 | 1313 |
| WildBench | 81.4% | 85.5% |
| LMArena Multi-Turn | 1339 | 1363 |
| LiveBench Language | 59.9% | — |
Frequently asked questions
Is Claude 3.7 Sonnet better than GPT-5 Mini?
GPT-5 Mini is the stronger model overall, scoring 41.7 to 39.4 on the Noometry Index.
Is Claude 3.7 Sonnet or GPT-5 Mini better for coding?
Claude 3.7 Sonnet scores higher on coding benchmarks: 41.1 versus 40.0 in the Noometry coding category.
How many benchmarks do Claude 3.7 Sonnet and GPT-5 Mini share?
40 benchmarks have published results for both models. Claude 3.7 Sonnet has 58 scored results on Noometry and GPT-5 Mini has 60.