Model comparison
Claude 3.7 Sonnet vs Gemini 3 Pro
Gemini 3 Pro is the stronger model overall, scoring 54.8 to 39.5 on the Noometry Index.
Last verified . 41 shared benchmarks.
GoogleSummary
- They share 41 benchmarks with published results for both. Claude 3.7 Sonnet scores higher in 1 category and Gemini 3 Pro in 9 categories; 10 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Gemini 3 Pro leads 52.5 to 18.6.
- The biggest single-benchmark swing is VPCT: 39% for Claude 3.7 Sonnet and 91% for Gemini 3 Pro.
Side by side
| Claude 3.7 Sonnet | Gemini 3 Pro | |
|---|---|---|
| Provider | Anthropic | |
| Noometry Index | 39.5 | 54.8 |
| Released | 2025-02-24 | 2025-11-18 |
| Weights | Proprietary | Proprietary |
| Context window | — | — |
| Max output | — | — |
| Input $ / M tokens | — | — |
| Output $ / M tokens | — | — |
| Results tracked | 58 | 67 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Gemini 3 Pro leads
Claude 3.7 Sonnet: 40.6 (#136), Gemini 3 Pro: 51.6 (#39)
| Benchmark | Claude 3.7 Sonnet | Gemini 3 Pro |
|---|---|---|
| SWE-bench Verified | 61% | 72.9% |
| SWE-bench Verified (bash only) | 52.8% | 74.2% |
| GSO | 3.8% | 18.6% |
| LMArena Coding | 1361 | 1481 |
| Aider Polyglot | 64.9% | — |
| LMArena WebDev | — | 1440 |
| SWE-bench Multilingual | — | 68.7% |
| WeirdML | — | 69.9% |
| LiveBench Coding | 74.5% | — |
| CadEval | 54% | — |
| ALE-Bench | — | 1,177 |
| AlgoTune | — | 1.83 |
Agentic & Tool Use Gemini 3 Pro leads
Claude 3.7 Sonnet: 34.1 (#50), Gemini 3 Pro: 40.6 (#23)
| Benchmark | Claude 3.7 Sonnet | Gemini 3 Pro |
|---|---|---|
| DeepResearch Bench | 43.6% | 46.3% |
| METR Time Horizons | 60% | 71% |
| Terminal-Bench | — | 69.4% |
| Berkeley Function Calling Leaderboard | — | 72.5% |
| GDPval | — | 40.3% |
| Remote Labor Index | — | 1.3% |
| TheAgentCompany | 30.9% | — |
| τ²-bench Airline | — | 80.5% |
| τ²-bench Banking | — | 18% |
| τ²-bench Retail | — | 75.9% |
| τ²-bench Telecom | — | 91% |
| Cybench | 20% | — |
| OSWorld | 35.8% | — |
| BALROG | — | 58.1% |
| LMArena Search | — | 1207 |
| Vending-Bench 2 | — | 5,478 |
Reasoning Gemini 3 Pro leads
Claude 3.7 Sonnet: 18.6 (#277), Gemini 3 Pro: 52.5 (#31)
| Benchmark | Claude 3.7 Sonnet | Gemini 3 Pro |
|---|---|---|
| ARC-AGI-2 | 0.9% | 31.1% |
| SimpleBench | 46.4% | 76.4% |
| ARC-AGI-1 | 28.6% | 75% |
| EnigmaEval | 4.2% | 18.2% |
| LMArena Hard Prompts | 1333 | 1480 |
| Epoch Capabilities Index | 141.16 | 152.92 |
| ForecastBench | 61.8 | 61.2 |
| Kagi LLM Benchmark | — | 80.1% |
| NYT Connections (extended) | — | 94.4% |
| CritPt | — | 6.9% |
| Chess Puzzles | — | 31% |
| LiveBench Reasoning | 87.8% | — |
| LiveBench Data Analysis | 74% | — |
| LiveBench | 76.1% | — |
Math Gemini 3 Pro leads
Claude 3.7 Sonnet: 37.5 (#153), Gemini 3 Pro: 49.9 (#59)
| Benchmark | Claude 3.7 Sonnet | Gemini 3 Pro |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 57.8% | 91.4% |
| Omni-MATH | 33% | 55.5% |
| LMArena Math | 1337 | 1476 |
| FrontierMath (Feb 2025 set) | 4.1% | 37.6% |
| MathArena Final-Answer Competitions | — | 67% |
| ProofBench | — | 20% |
| LiveBench Math | 79% | — |
| MATH Level 5 | 91.2% | — |
| FrontierMath Tier 4 (v1) | — | 18.8% |
Knowledge Gemini 3 Pro leads
Claude 3.7 Sonnet: 39.8 (#130), Gemini 3 Pro: 64.4 (#16)
| Benchmark | Claude 3.7 Sonnet | Gemini 3 Pro |
|---|---|---|
| GPQA Diamond | 79.7% | 92.6% |
| Humanity's Last Exam | 8% | 37.5% |
| MMLU-Pro | 78.4% | 90.3% |
| GPQA (HELM) | 60.8% | 80.3% |
| LMArena Expert | 1321 | 1475 |
| Confabulations | 14.7% | — |
| Vectara Hallucination Rate | — | 13.6% |
Multimodal Gemini 3 Pro leads
Claude 3.7 Sonnet: 33.7 (#95), Gemini 3 Pro: 57.6 (#2)
| Benchmark | Claude 3.7 Sonnet | Gemini 3 Pro |
|---|---|---|
| LMArena Vision | 1169 | 1305 |
| GeoBench | 68% | 84% |
| VPCT | 39% | 91% |
| LMArena Document | — | 1434 |
| SpatialViz-Bench | 33.9% | — |
Multilingual Gemini 3 Pro leads
Claude 3.7 Sonnet: 44.1 (#179), Gemini 3 Pro: 56.9 (#16)
| Benchmark | Claude 3.7 Sonnet | Gemini 3 Pro |
|---|---|---|
| LMArena Non-English | 1296 | 1474 |
| LMArena Chinese | 1299 | 1523 |
| LMArena French | 1303 | 1492 |
| LMArena German | 1301 | 1515 |
| LMArena Japanese | 1267 | 1510 |
| LMArena Korean | 1249 | 1448 |
| LMArena Russian | 1311 | 1493 |
| LMArena Spanish | 1298 | 1470 |
Instruction Following Gemini 3 Pro leads
Claude 3.7 Sonnet: 72.9 (#125), Gemini 3 Pro: 76.3 (#45)
| Benchmark | Claude 3.7 Sonnet | Gemini 3 Pro |
|---|---|---|
| IFEval | 83.4% | 87.7% |
| LMArena Instruction Following | 1352 | 1458 |
| LiveBench Instruction Following | 81.3% | — |
Long Context Claude 3.7 Sonnet leads
Claude 3.7 Sonnet: 50.3 (#10), Gemini 3 Pro: 44.0 (#79)
| Benchmark | Claude 3.7 Sonnet | Gemini 3 Pro |
|---|---|---|
| LMArena Longer Query | 1373 | 1471 |
| Fiction.LiveBench | 83.3% | — |
| CL-bench | — | 15.8% |
Writing & Preference Gemini 3 Pro leads
Claude 3.7 Sonnet: 54.4 (#150), Gemini 3 Pro: 66.4 (#35)
| Benchmark | Claude 3.7 Sonnet | Gemini 3 Pro |
|---|---|---|
| LMArena Text | 1314 | 1479 |
| LMArena Creative Writing | 1332 | 1482 |
| EQ-Bench Creative Writing | 1412 | 1525 |
| WildBench | 81.4% | 85.9% |
| LMArena Multi-Turn | 1339 | 1484 |
| Short-Story Creative Writing | 81.1% | — |
| LiveBench Language | 59.9% | — |
Frequently asked questions
Is Claude 3.7 Sonnet better than Gemini 3 Pro?
Gemini 3 Pro is the stronger model overall, scoring 54.8 to 39.5 on the Noometry Index.
Is Claude 3.7 Sonnet or Gemini 3 Pro better for coding?
Gemini 3 Pro scores higher on coding benchmarks: 51.6 versus 40.6 in the Noometry coding category.
How many benchmarks do Claude 3.7 Sonnet and Gemini 3 Pro share?
41 benchmarks have published results for both models. Claude 3.7 Sonnet has 58 scored results on Noometry and Gemini 3 Pro has 67.