# Claude 3.5 Sonnet vs Gemini 3 Pro

> Gemini 3 Pro is the stronger model overall, scoring 54.7 to 34.4 on the Noometry Index.

- Canonical page: https://noometry.com/compare/claude-3-5-sonnet-vs-gemini-3-pro
- Last updated: 2026-10-10
- Shared benchmarks: 39

## Summary

- They share 39 benchmarks with published results for both. Claude 3.5 Sonnet scores higher in 0 categories and Gemini 3 Pro in 10 categories; 10 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Gemini 3 Pro leads 64.3 to 28.5.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 8.5% for Claude 3.5 Sonnet and 91.4% for Gemini 3 Pro.

## Snapshot

| | Claude 3.5 Sonnet | Gemini 3 Pro |
|---|---|---|
| Provider | Anthropic | Google |
| Noometry Index | 34.4 | 54.7 |
| Rank | 225 | 23 |
| Context | — | — |
| Input $/M | — | — |
| Output $/M | — | — |
| Weights | Proprietary | Proprietary |

## Coding

- Claude 3.5 Sonnet: 39.5 (#145)
- Gemini 3 Pro: 52.1 (#33)

| Benchmark | Claude 3.5 Sonnet | Gemini 3 Pro |
|---|---|---|
| GSO | 4.6% | 18.6% |
| WeirdML | 40% | 69.9% |
| LMArena Coding | 1343 | 1482 |
| SWE-bench Verified | — | 72.9% |
| SWE-bench Verified (bash only) | — | 74.2% |
| Aider Polyglot | 51.6% | — |
| LMArena WebDev | — | 1439 |
| SWE-bench Multilingual | — | 68.7% |
| BigCodeBench Instruct | 46.8% | — |
| LiveBench Coding | 67.1% | — |
| BigCodeBench Complete | 58.6% | — |
| CadEval | 48% | — |
| ALE-Bench | — | 1,177 |
| AlgoTune | — | 1.83 |
| HumanEval+ | 81.7% | — |
| MBPP+ | 74.3% | — |

## Agentic & Tool Use

- Claude 3.5 Sonnet: 32.1 (#61)
- Gemini 3 Pro: 40.5 (#21)

| Benchmark | Claude 3.5 Sonnet | Gemini 3 Pro |
|---|---|---|
| BALROG | 32.6% | 58.1% |
| METR Time Horizons | 45.2% | 71% |
| Terminal-Bench | — | 69.4% |
| Berkeley Function Calling Leaderboard | — | 72.5% |
| GDPval | — | 40.3% |
| Remote Labor Index | — | 1.3% |
| TheAgentCompany | 24% | — |
| τ²-bench Airline | — | 80.5% |
| τ²-bench Banking | — | 18% |
| τ²-bench Retail | — | 75.9% |
| τ²-bench Telecom | — | 91% |
| Cybench | 17.5% | — |
| DeepResearch Bench | — | 46.3% |
| LMArena Search | — | 1207 |
| Vending-Bench 2 | — | 5,478 |

## Reasoning

- Claude 3.5 Sonnet: 22.5 (#176)
- Gemini 3 Pro: 52.5 (#28)

| Benchmark | Claude 3.5 Sonnet | Gemini 3 Pro |
|---|---|---|
| SimpleBench | 41.4% | 76.4% |
| EnigmaEval | 0.9% | 18.2% |
| LMArena Hard Prompts | 1305 | 1481 |
| Epoch Capabilities Index | 133.55 | 152.96 |
| ForecastBench | 60.7 | 61.2 |
| ARC-AGI-2 | — | 31.1% |
| Kagi LLM Benchmark | — | 80.1% |
| NYT Connections (extended) | — | 94.4% |
| ARC-AGI-1 | — | 75% |
| CritPt | — | 6.9% |
| Chess Puzzles | — | 31% |
| LiveBench Reasoning | 56.7% | — |
| DTBench | 67.8% | — |
| LiveBench Data Analysis | 55% | — |
| LiveBench | 59% | — |

## Math

- Claude 3.5 Sonnet: 18.0 (#282)
- Gemini 3 Pro: 48.8 (#50)

| Benchmark | Claude 3.5 Sonnet | Gemini 3 Pro |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 8.5% | 91.4% |
| Omni-MATH | 27.6% | 55.5% |
| LMArena Math | 1306 | 1475 |
| FrontierMath (Feb 2025 set) | 2.1% | 37.6% |
| FrontierMath Tier 4 (v1) | 0% | 18.8% |
| MathArena Final-Answer Competitions | — | 67% |
| ProofBench | — | 20% |
| LiveBench Math | 52.3% | — |
| MATH Level 5 | 56.9% | — |

## Knowledge

- Claude 3.5 Sonnet: 28.5 (#236)
- Gemini 3 Pro: 64.3 (#15)

| Benchmark | Claude 3.5 Sonnet | Gemini 3 Pro |
|---|---|---|
| GPQA Diamond | 55.3% | 92.6% |
| Humanity's Last Exam | 4.1% | 37.5% |
| MMLU-Pro | 77.7% | 90.3% |
| GPQA (HELM) | 56.5% | 80.3% |
| LMArena Expert | 1264 | 1475 |
| Confabulations | 19.9% | — |
| Vectara Hallucination Rate | — | 13.6% |
| MMLU | 87.3% | — |

## Multimodal

- Claude 3.5 Sonnet: 26.0 (#115)
- Gemini 3 Pro: 56.9 (#1)

| Benchmark | Claude 3.5 Sonnet | Gemini 3 Pro |
|---|---|---|
| LMArena Vision | 1125 | 1305 |
| GeoBench | 62% | 84% |
| VPCT | 33% | 91% |
| Video-MME | 60% | — |
| LMArena Document | — | 1434 |

## Multilingual

- Claude 3.5 Sonnet: 43.4 (#180)
- Gemini 3 Pro: 57.1 (#14)

| Benchmark | Claude 3.5 Sonnet | Gemini 3 Pro |
|---|---|---|
| LMArena Non-English | 1283 | 1474 |
| LMArena Chinese | 1271 | 1522 |
| LMArena French | 1304 | 1492 |
| LMArena German | 1297 | 1517 |
| LMArena Japanese | 1234 | 1513 |
| LMArena Korean | 1200 | 1449 |
| LMArena Russian | 1305 | 1493 |
| LMArena Spanish | 1289 | 1468 |

## Instruction Following

- Claude 3.5 Sonnet: 69.0 (#177)
- Gemini 3 Pro: 76.4 (#42)

| Benchmark | Claude 3.5 Sonnet | Gemini 3 Pro |
|---|---|---|
| IFEval | 85.5% | 87.7% |
| LMArena Instruction Following | 1297 | 1458 |
| LiveBench Instruction Following | 69.3% | — |

## Long Context

- Claude 3.5 Sonnet: 39.9 (#162)
- Gemini 3 Pro: 44.0 (#75)

| Benchmark | Claude 3.5 Sonnet | Gemini 3 Pro |
|---|---|---|
| LMArena Longer Query | 1311 | 1472 |
| CL-bench | — | 15.8% |

## Writing & Preference

- Claude 3.5 Sonnet: 53.0 (#159)
- Gemini 3 Pro: 66.6 (#34)

| Benchmark | Claude 3.5 Sonnet | Gemini 3 Pro |
|---|---|---|
| LMArena Text | 1298 | 1479 |
| LMArena Creative Writing | 1292 | 1482 |
| EQ-Bench Creative Writing | 1451 | 1525 |
| WildBench | 79.2% | 85.9% |
| LMArena Multi-Turn | 1326 | 1484 |
| Short-Story Creative Writing | 80.3% | — |
| LiveBench Language | 53.8% | — |

## FAQ

### Is Claude 3.5 Sonnet better than Gemini 3 Pro?

Gemini 3 Pro is the stronger model overall, scoring 54.7 to 34.4 on the Noometry Index.

### Is Claude 3.5 Sonnet or Gemini 3 Pro better for coding?

Gemini 3 Pro scores higher on coding benchmarks: 52.1 versus 39.5 in the Noometry coding category.

### How many benchmarks do Claude 3.5 Sonnet and Gemini 3 Pro share?

39 benchmarks have published results for both models. Claude 3.5 Sonnet has 60 scored results on Noometry and Gemini 3 Pro has 67.