Model comparison
Claude 3.7 Sonnet vs Gemini 2.5 Flash
Claude 3.7 Sonnet and Gemini 2.5 Flash score almost the same on the Noometry Index (39.5 vs 39.3), so choose on price, context window or the category you care about most.
Last verified . 42 shared benchmarks.
GoogleSummary
- They share 42 benchmarks with published results for both. Claude 3.7 Sonnet scores higher in 6 categories and Gemini 2.5 Flash in 4 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in multilingual, where Gemini 2.5 Flash leads 52.3 to 44.1.
- The biggest single-benchmark swing is SWE-bench Verified (bash only): 52.8% for Claude 3.7 Sonnet and 28.7% for Gemini 2.5 Flash.
Side by side
| Claude 3.7 Sonnet | Gemini 2.5 Flash | |
|---|---|---|
| Provider | Anthropic | |
| Noometry Index | 39.5 | 39.3 |
| Released | 2025-02-24 | 2025-04-17 |
| Weights | Proprietary | Proprietary |
| Context window | — | 1.05M |
| Max output | — | 66K |
| Input $ / M tokens | — | $0.30 |
| Output $ / M tokens | — | $2.50 |
| Results tracked | 58 | 54 |
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), Gemini 2.5 Flash: 35.8 (#220)
| Benchmark | Claude 3.7 Sonnet | Gemini 2.5 Flash |
|---|---|---|
| SWE-bench Verified (bash only) | 52.8% | 28.7% |
| Aider Polyglot | 64.9% | 55.1% |
| LMArena Coding | 1361 | 1424 |
| SWE-bench Verified | 61% | — |
| GSO | 3.8% | — |
| WeirdML | — | 41.9% |
| LiveBench Coding | 74.5% | — |
| CadEval | 54% | — |
| ALE-Bench | — | 661.88 |
Agentic & Tool Use Claude 3.7 Sonnet leads
Claude 3.7 Sonnet: 34.1 (#50), Gemini 2.5 Flash: 30.8 (#74)
| Benchmark | Claude 3.7 Sonnet | Gemini 2.5 Flash |
|---|---|---|
| TheAgentCompany | 30.9% | 41.1% |
| Terminal-Bench | — | 17.1% |
| Berkeley Function Calling Leaderboard | — | 56.2% |
| Cybench | 20% | — |
| DeepResearch Bench | 43.6% | — |
| OSWorld | 35.8% | — |
| BALROG | — | 33.5% |
| METR Time Horizons | 60% | — |
| Vending-Bench 2 | — | 548.84 |
Reasoning Too close to call
Claude 3.7 Sonnet: 18.6 (#277), Gemini 2.5 Flash: 18.1 (#286)
| Benchmark | Claude 3.7 Sonnet | Gemini 2.5 Flash |
|---|---|---|
| ARC-AGI-2 | 0.9% | 2.5% |
| SimpleBench | 46.4% | 41.2% |
| ARC-AGI-1 | 28.6% | 33.3% |
| EnigmaEval | 4.2% | 2.7% |
| LMArena Hard Prompts | 1333 | 1422 |
| Epoch Capabilities Index | 141.16 | 143.03 |
| ForecastBench | 61.8 | 60.6 |
| Kagi LLM Benchmark | — | 56.8% |
| CritPt | — | 1.1% |
| LiveBench Reasoning | 87.8% | — |
| DTBench | — | 76.5% |
| LiveBench Data Analysis | 74% | — |
| LMCA | — | 27.5% |
| LiveBench | 76.1% | — |
Math Gemini 2.5 Flash leads
Claude 3.7 Sonnet: 37.5 (#153), Gemini 2.5 Flash: 39.9 (#98)
| Benchmark | Claude 3.7 Sonnet | Gemini 2.5 Flash |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 57.8% | 73.1% |
| Omni-MATH | 33% | 38.5% |
| LMArena Math | 1337 | 1415 |
| FrontierMath (Feb 2025 set) | 4.1% | 4.8% |
| LiveBench Math | 79% | — |
| MATH Level 5 | 91.2% | — |
| FrontierMath Tier 4 (v1) | — | 4.2% |
Knowledge Claude 3.7 Sonnet leads
Claude 3.7 Sonnet: 39.8 (#130), Gemini 2.5 Flash: 36.4 (#168)
| Benchmark | Claude 3.7 Sonnet | Gemini 2.5 Flash |
|---|---|---|
| Humanity's Last Exam | 8% | 12.1% |
| MMLU-Pro | 78.4% | 63.9% |
| Confabulations | 14.7% | 16.8% |
| GPQA (HELM) | 60.8% | 39% |
| LMArena Expert | 1321 | 1426 |
| GPQA Diamond | 79.7% | — |
| Vectara Hallucination Rate | — | 7.8% |
Multimodal Gemini 2.5 Flash leads
Claude 3.7 Sonnet: 33.7 (#95), Gemini 2.5 Flash: 41.8 (#32)
| Benchmark | Claude 3.7 Sonnet | Gemini 2.5 Flash |
|---|---|---|
| LMArena Vision | 1169 | 1253 |
| GeoBench | 68% | 76% |
| VPCT | 39% | 46.2% |
| SpatialViz-Bench | 33.9% | 36.9% |
Multilingual Gemini 2.5 Flash leads
Claude 3.7 Sonnet: 44.1 (#179), Gemini 2.5 Flash: 52.3 (#88)
| Benchmark | Claude 3.7 Sonnet | Gemini 2.5 Flash |
|---|---|---|
| LMArena Non-English | 1296 | 1409 |
| LMArena Chinese | 1299 | 1450 |
| LMArena French | 1303 | 1433 |
| LMArena German | 1301 | 1418 |
| LMArena Japanese | 1267 | 1405 |
| LMArena Korean | 1249 | 1385 |
| LMArena Russian | 1311 | 1415 |
| LMArena Spanish | 1298 | 1421 |
Instruction Following Gemini 2.5 Flash leads
Claude 3.7 Sonnet: 72.9 (#125), Gemini 2.5 Flash: 75.7 (#54)
| Benchmark | Claude 3.7 Sonnet | Gemini 2.5 Flash |
|---|---|---|
| IFEval | 83.4% | 89.8% |
| LMArena Instruction Following | 1352 | 1405 |
| LiveBench Instruction Following | 81.3% | — |
Long Context Claude 3.7 Sonnet leads
Claude 3.7 Sonnet: 50.3 (#10), Gemini 2.5 Flash: 47.5 (#17)
| Benchmark | Claude 3.7 Sonnet | Gemini 2.5 Flash |
|---|---|---|
| Fiction.LiveBench | 83.3% | 77.8% |
| LMArena Longer Query | 1373 | 1419 |
Writing & Preference Too close to call
Claude 3.7 Sonnet: 54.4 (#150), Gemini 2.5 Flash: 53.8 (#157)
| Benchmark | Claude 3.7 Sonnet | Gemini 2.5 Flash |
|---|---|---|
| LMArena Text | 1314 | 1417 |
| LMArena Creative Writing | 1332 | 1400 |
| Short-Story Creative Writing | 81.1% | 76.5% |
| EQ-Bench Creative Writing | 1412 | 1137 |
| WildBench | 81.4% | 81.7% |
| LMArena Multi-Turn | 1339 | 1408 |
| LiveBench Language | 59.9% | — |
Frequently asked questions
Is Claude 3.7 Sonnet better than Gemini 2.5 Flash?
Claude 3.7 Sonnet and Gemini 2.5 Flash score almost the same on the Noometry Index (39.5 vs 39.3), so choose on price, context window or the category you care about most.
Is Claude 3.7 Sonnet or Gemini 2.5 Flash better for coding?
Claude 3.7 Sonnet scores higher on coding benchmarks: 40.6 versus 35.8 in the Noometry coding category.
How many benchmarks do Claude 3.7 Sonnet and Gemini 2.5 Flash share?
42 benchmarks have published results for both models. Claude 3.7 Sonnet has 58 scored results on Noometry and Gemini 2.5 Flash has 54.