# Claude 3 Sonnet vs Gemini 3.1 Pro Preview

> Gemini 3.1 Pro Preview is the stronger model overall, scoring 56.7 to 28.8 on the Noometry Index.

- Canonical page: https://noometry.com/compare/claude-3-sonnet-vs-gemini-3-1-pro-preview
- Last updated: 2026-10-10
- Shared benchmarks: 23

## Summary

- They share 23 benchmarks with published results for both. Claude 3 Sonnet scores higher in 0 categories and Gemini 3.1 Pro Preview in 9 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Gemini 3.1 Pro Preview leads 71.7 to 19.7.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 2.5% for Claude 3 Sonnet and 95.6% for Gemini 3.1 Pro Preview.

## Snapshot

| | Claude 3 Sonnet | Gemini 3.1 Pro Preview |
|---|---|---|
| Provider | Anthropic | Google |
| Noometry Index | 28.8 | 56.7 |
| Rank | 311 | 20 |
| Context | — | 1.05M |
| Input $/M | — | $2 |
| Output $/M | — | $12 |
| Weights | Proprietary | Proprietary |

## Coding

- Claude 3 Sonnet: 29.2 (#296)
- Gemini 3.1 Pro Preview: 42.7 (#88)

| Benchmark | Claude 3 Sonnet | Gemini 3.1 Pro Preview |
|---|---|---|
| WeirdML | 10.2% | 72.1% |
| LMArena Coding | 1223 | 1483 |
| SWE-bench Verified | — | 75.6% |
| DeepSWE | — | 11.7% |
| LMArena WebDev | — | 1446 |
| SciCode | — | 58.9% |
| GSO | — | 22.6% |
| BigCodeBench Instruct | 42.7% | — |
| MirrorCode | — | 8.9% |
| BigCodeBench Complete | 53.8% | — |
| ALE-Bench | — | 1,161 |
| AlgoTune | — | 2.02 |
| HumanEval+ | 64% | — |
| MBPP+ | 69.3% | — |

## Agentic & Tool Use

- Claude 3 Sonnet: —
- Gemini 3.1 Pro Preview: 37.7 (#33)

| Benchmark | Claude 3 Sonnet | Gemini 3.1 Pro Preview |
|---|---|---|
| Terminal-Bench | — | 80.2% |
| APEX-Agents | — | 35.3% |
| τ²-bench Banking | — | 26% |
| DeepResearch Bench | — | 47.8% |
| PostTrainBench | — | 22% |
| BALROG | — | 57% |
| ExploitBench | — | 26.1% |
| GBAEval | — | 0.8% |
| GDP.pdf | — | 17% |
| LMArena Search | — | 1211 |
| METR Time Horizons | — | 77% |
| Vending-Bench 2 | — | 3,774 |

## Reasoning

- Claude 3 Sonnet: 19.7 (#232)
- Gemini 3.1 Pro Preview: 71.7 (#9)

| Benchmark | Claude 3 Sonnet | Gemini 3.1 Pro Preview |
|---|---|---|
| LMArena Hard Prompts | 1197 | 1484 |
| DTBench | 53.6% | 97.1% |
| Epoch Capabilities Index | 120.69 | 154.92 |
| ARC-AGI-2 | — | 77.1% |
| SimpleBench | — | 79.6% |
| NYT Connections (extended) | — | 97.4% |
| ARC-AGI-1 | — | 98% |
| CritPt | — | 17.7% |
| Chess Puzzles | — | 55% |
| EnigmaEval | — | 36.8% |
| Thematic Generalization | — | 79.4% |
| EBR-Bench | — | 14.3% |
| Mystery Game Puzzles | — | 34% |
| LMCA | — | 53.8% |
| ForecastBench | — | 59 |
| WinoGrande | 75.1% | — |

## Math

- Claude 3 Sonnet: 10.0 (#305)
- Gemini 3.1 Pro Preview: 61.8 (#28)

| Benchmark | Claude 3 Sonnet | Gemini 3.1 Pro Preview |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 2.5% | 95.6% |
| LMArena Math | 1213 | 1485 |
| FrontierMath (Tiers 1-3) | — | 59.6% |
| FrontierMath Tier 4 | — | 26.8% |
| MathArena Final-Answer Competitions | — | 86.5% |
| ProofBench | — | 26% |
| MATH Level 5 | 18.2% | — |
| FrontierMath (Feb 2025 set) | — | 36.9% |
| FrontierMath Tier 4 (v1) | — | 16.7% |

## Knowledge

- Claude 3 Sonnet: 21.1 (#271)
- Gemini 3.1 Pro Preview: 71.8 (#3)

| Benchmark | Claude 3 Sonnet | Gemini 3.1 Pro Preview |
|---|---|---|
| GPQA Diamond | 40.6% | 94.4% |
| LMArena Expert | 1173 | 1484 |
| Humanity's Last Exam | — | 46.4% |
| SimpleQA Verified | — | 73.5% |
| Vectara Hallucination Rate | — | 10.4% |
| MMLU | 75.9% | — |

## Multimodal

- Claude 3 Sonnet: 24.7 (#120)
- Gemini 3.1 Pro Preview: 38.1 (#60)

| Benchmark | Claude 3 Sonnet | Gemini 3.1 Pro Preview |
|---|---|---|
| LMArena Vision | 984 | 1296 |
| Blueprint-Bench 2 | — | 26.5% |
| Furniture Assembly | — | 26.7% |
| LMArena Document | — | 1444 |

## Multilingual

- Claude 3 Sonnet: 38.0 (#229)
- Gemini 3.1 Pro Preview: 57.1 (#11)

| Benchmark | Claude 3 Sonnet | Gemini 3.1 Pro Preview |
|---|---|---|
| LMArena Non-English | 1205 | 1475 |
| LMArena Chinese | 1188 | 1528 |
| LMArena French | 1228 | 1488 |
| LMArena German | 1204 | 1491 |
| LMArena Japanese | 1131 | 1494 |
| LMArena Korean | 1128 | 1458 |
| LMArena Russian | 1227 | 1496 |
| LMArena Spanish | 1204 | 1480 |

## Instruction Following

- Claude 3 Sonnet: 63.1 (#230)
- Gemini 3.1 Pro Preview: 77.1 (#31)

| Benchmark | Claude 3 Sonnet | Gemini 3.1 Pro Preview |
|---|---|---|
| LMArena Instruction Following | 1200 | 1465 |

## Long Context

- Claude 3 Sonnet: 36.7 (#223)
- Gemini 3.1 Pro Preview: 47.3 (#17)

| Benchmark | Claude 3 Sonnet | Gemini 3.1 Pro Preview |
|---|---|---|
| LMArena Longer Query | 1211 | 1482 |
| CL-bench | — | 20.8% |
| CL-bench Life | — | 16.9% |

## Writing & Preference

- Claude 3 Sonnet: 42.3 (#233)
- Gemini 3.1 Pro Preview: 66.3 (#36)

| Benchmark | Claude 3 Sonnet | Gemini 3.1 Pro Preview |
|---|---|---|
| LMArena Text | 1218 | 1480 |
| LMArena Creative Writing | 1186 | 1482 |
| LMArena Multi-Turn | 1227 | 1486 |
| EQ-Bench Creative Writing | — | 1491 |
| EQ-Bench 4 | — | 1142 |

## FAQ

### Is Claude 3 Sonnet better than Gemini 3.1 Pro Preview?

Gemini 3.1 Pro Preview is the stronger model overall, scoring 56.7 to 28.8 on the Noometry Index.

### Is Claude 3 Sonnet or Gemini 3.1 Pro Preview better for coding?

Gemini 3.1 Pro Preview scores higher on coding benchmarks: 42.7 versus 29.2 in the Noometry coding category.

### How many benchmarks do Claude 3 Sonnet and Gemini 3.1 Pro Preview share?

23 benchmarks have published results for both models. Claude 3 Sonnet has 30 scored results on Noometry and Gemini 3.1 Pro Preview has 71.