Model comparison
Claude 3.5 Sonnet vs Gemini 2.5 Pro
Gemini 2.5 Pro is the stronger model overall, scoring 44.9 to 34.4 on the Noometry Index.
Last verified . 53 shared benchmarks.
GoogleSummary
- They share 53 benchmarks with published results for both. Claude 3.5 Sonnet scores higher in 1 category and Gemini 2.5 Pro in 9 categories; 10 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Gemini 2.5 Pro leads 55.8 to 28.5.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 8.5% for Claude 3.5 Sonnet and 84.7% for Gemini 2.5 Pro.
Side by side
| Claude 3.5 Sonnet | Gemini 2.5 Pro | |
|---|---|---|
| Provider | Anthropic | |
| Noometry Index | 34.4 | 44.9 |
| Released | 2024-06-20 | 2025-03-25 |
| Weights | Proprietary | Proprietary |
| Context window | — | 1.05M |
| Max output | — | 66K |
| Input $ / M tokens | — | $1.25 |
| Output $ / M tokens | — | $10 |
| Results tracked | 60 | 78 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Gemini 2.5 Pro leads
Claude 3.5 Sonnet: 39.5 (#145), Gemini 2.5 Pro: 42.8 (#86)
| Benchmark | Claude 3.5 Sonnet | Gemini 2.5 Pro |
|---|---|---|
| Aider Polyglot | 51.6% | 83.1% |
| GSO | 4.6% | 3.9% |
| WeirdML | 40% | 54% |
| LiveBench Coding | 67.1% | 85.9% |
| LMArena Coding | 1343 | 1452 |
| CadEval | 48% | 64% |
| SWE-bench Verified | — | 57.6% |
| SWE-bench Verified (bash only) | — | 53.6% |
| LMArena WebDev | — | 1227 |
| SciCode | — | 42.8% |
| BigCodeBench Instruct | 46.8% | — |
| BigCodeBench Complete | 58.6% | — |
| ALE-Bench | — | 785.52 |
| AlgoTune | — | 1.51 |
| HumanEval+ | 81.7% | — |
| MBPP+ | 74.3% | — |
Agentic & Tool Use Claude 3.5 Sonnet leads
Claude 3.5 Sonnet: 32.1 (#61), Gemini 2.5 Pro: 29.1 (#81)
| Benchmark | Claude 3.5 Sonnet | Gemini 2.5 Pro |
|---|---|---|
| TheAgentCompany | 24% | 30.3% |
| BALROG | 32.6% | 43.3% |
| METR Time Horizons | 45.2% | 55.4% |
| Terminal-Bench | — | 32.6% |
| GDPval | — | 23.3% |
| Remote Labor Index | — | 0.8% |
| τ²-bench Banking | — | 13.7% |
| Cybench | 17.5% | — |
| DeepResearch Bench | — | 42.8% |
| LMArena Search | — | 1142 |
| Vending-Bench 2 | — | 573.64 |
Reasoning Gemini 2.5 Pro leads
Claude 3.5 Sonnet: 22.5 (#176), Gemini 2.5 Pro: 28.8 (#85)
| Benchmark | Claude 3.5 Sonnet | Gemini 2.5 Pro |
|---|---|---|
| SimpleBench | 41.4% | 62.4% |
| EnigmaEval | 0.9% | 5.6% |
| LiveBench Reasoning | 56.7% | 89.8% |
| LMArena Hard Prompts | 1305 | 1455 |
| DTBench | 67.8% | 82.4% |
| LiveBench Data Analysis | 55% | 79.9% |
| Epoch Capabilities Index | 133.55 | 145.32 |
| ForecastBench | 60.7 | 61.3 |
| LiveBench | 59% | 82.3% |
| ARC-AGI-2 | — | 4.9% |
| Kagi LLM Benchmark | — | 70.3% |
| ARC-AGI-1 | — | 41% |
| CritPt | — | 2% |
| Chess Puzzles | — | 20% |
| LMCA | — | 34.8% |
Math Gemini 2.5 Pro leads
Claude 3.5 Sonnet: 18.0 (#282), Gemini 2.5 Pro: 31.9 (#191)
| Benchmark | Claude 3.5 Sonnet | Gemini 2.5 Pro |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 8.5% | 84.7% |
| Omni-MATH | 27.6% | 41.6% |
| LiveBench Math | 52.3% | 90.2% |
| LMArena Math | 1306 | 1450 |
| MATH Level 5 | 56.9% | 95.9% |
| FrontierMath (Feb 2025 set) | 2.1% | 14.1% |
| FrontierMath Tier 4 (v1) | 0% | 4.2% |
| FrontierMath (Tiers 1-3) | — | 24.6% |
| FrontierMath Tier 4 | — | 0% |
Knowledge Gemini 2.5 Pro leads
Claude 3.5 Sonnet: 28.5 (#236), Gemini 2.5 Pro: 55.8 (#41)
| Benchmark | Claude 3.5 Sonnet | Gemini 2.5 Pro |
|---|---|---|
| GPQA Diamond | 55.3% | 85.3% |
| Humanity's Last Exam | 4.1% | 21.6% |
| MMLU-Pro | 77.7% | 86.3% |
| Confabulations | 19.9% | 10.6% |
| GPQA (HELM) | 56.5% | 74.9% |
| LMArena Expert | 1264 | 1452 |
| Vectara Hallucination Rate | — | 7% |
| MMLU | 87.3% | — |
Multimodal Gemini 2.5 Pro leads
Claude 3.5 Sonnet: 26.0 (#115), Gemini 2.5 Pro: 44.5 (#16)
| Benchmark | Claude 3.5 Sonnet | Gemini 2.5 Pro |
|---|---|---|
| LMArena Vision | 1125 | 1263 |
| GeoBench | 62% | 86% |
| VPCT | 33% | 48% |
| Video-MME | 60% | — |
| LMArena Document | — | 1421 |
| SpatialViz-Bench | — | 44.7% |
Multilingual Gemini 2.5 Pro leads
Claude 3.5 Sonnet: 43.4 (#180), Gemini 2.5 Pro: 55.5 (#30)
| Benchmark | Claude 3.5 Sonnet | Gemini 2.5 Pro |
|---|---|---|
| LMArena Non-English | 1283 | 1452 |
| LMArena Chinese | 1271 | 1507 |
| LMArena French | 1304 | 1473 |
| LMArena German | 1297 | 1487 |
| LMArena Japanese | 1234 | 1462 |
| LMArena Korean | 1200 | 1435 |
| LMArena Russian | 1305 | 1460 |
| LMArena Spanish | 1289 | 1473 |
Instruction Following Gemini 2.5 Pro leads
Claude 3.5 Sonnet: 69.0 (#177), Gemini 2.5 Pro: 75.2 (#71)
| Benchmark | Claude 3.5 Sonnet | Gemini 2.5 Pro |
|---|---|---|
| LiveBench Instruction Following | 69.3% | 80.6% |
| IFEval | 85.5% | 84% |
| LMArena Instruction Following | 1297 | 1436 |
Long Context Gemini 2.5 Pro leads
Claude 3.5 Sonnet: 39.9 (#162), Gemini 2.5 Pro: 59.7 (#5)
| Benchmark | Claude 3.5 Sonnet | Gemini 2.5 Pro |
|---|---|---|
| LMArena Longer Query | 1311 | 1449 |
| Fiction.LiveBench | — | 91.7% |
Writing & Preference Gemini 2.5 Pro leads
Claude 3.5 Sonnet: 53.0 (#159), Gemini 2.5 Pro: 63.9 (#60)
| Benchmark | Claude 3.5 Sonnet | Gemini 2.5 Pro |
|---|---|---|
| LMArena Text | 1298 | 1458 |
| LMArena Creative Writing | 1292 | 1453 |
| Short-Story Creative Writing | 80.3% | 83.8% |
| EQ-Bench Creative Writing | 1451 | 1421 |
| WildBench | 79.2% | 85.7% |
| LMArena Multi-Turn | 1326 | 1453 |
| LiveBench Language | 53.8% | 67.8% |
Frequently asked questions
Is Claude 3.5 Sonnet better than Gemini 2.5 Pro?
Gemini 2.5 Pro is the stronger model overall, scoring 44.9 to 34.4 on the Noometry Index.
Is Claude 3.5 Sonnet or Gemini 2.5 Pro better for coding?
Gemini 2.5 Pro scores higher on coding benchmarks: 42.8 versus 39.5 in the Noometry coding category.
How many benchmarks do Claude 3.5 Sonnet and Gemini 2.5 Pro share?
53 benchmarks have published results for both models. Claude 3.5 Sonnet has 60 scored results on Noometry and Gemini 2.5 Pro has 78.