# Claude Sonnet 5 vs Gemini 2.5 Pro

> Claude Sonnet 5 is the stronger model overall, scoring 54.6 to 45.0 on the Noometry Index.

- Canonical page: https://noometry.com/compare/claude-sonnet-5-vs-gemini-2-5-pro
- Last updated: 2026-10-10
- Shared benchmarks: 38

## Summary

- They share 38 benchmarks with published results for both. Claude Sonnet 5 scores higher in 6 categories and Gemini 2.5 Pro in 4 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in math, where Claude Sonnet 5 leads 66.2 to 32.5.
- The biggest single-benchmark swing is FrontierMath (Tiers 1-3): 65.6% for Claude Sonnet 5 and 24.6% for Gemini 2.5 Pro.
- Gemini 2.5 Pro is cheaper at $1.25 / $10 per million input/output tokens, against $2 / $10 for Claude Sonnet 5.
- Gemini 2.5 Pro accepts more context: 1.05M tokens versus 1M.

## Snapshot

| | Claude Sonnet 5 | Gemini 2.5 Pro |
|---|---|---|
| Provider | Anthropic | Google |
| Noometry Index | 54.6 | 45.0 |
| Rank | 29 | 75 |
| Context | 1M | 1.05M |
| Input $/M | $2 | $1.25 |
| Output $/M | $10 | $10 |
| Weights | Proprietary | Proprietary |

## Coding

- Claude Sonnet 5: 55.5 (#26)
- Gemini 2.5 Pro: 42.4 (#101)

| Benchmark | Claude Sonnet 5 | Gemini 2.5 Pro |
|---|---|---|
| LMArena WebDev | 1541 | 1227 |
| SciCode | 54.3% | 42.8% |
| GSO | 37.3% | 3.9% |
| WeirdML | 68.8% | 54% |
| LMArena Coding | 1483 | 1452 |
| ALE-Bench | 1,463 | 785.52 |
| SWE-bench Verified | — | 57.6% |
| DeepSWE | 53.8% | — |
| FrontierCode | 42.7% | — |
| SWE-bench Verified (bash only) | — | 53.6% |
| Aider Polyglot | — | 83.1% |
| CursorBench | 34.1% | — |
| LiveBench Coding | — | 85.9% |
| CadEval | — | 64% |
| AlgoTune | — | 1.51 |

## Agentic & Tool Use

- Claude Sonnet 5: 42.8 (#18)
- Gemini 2.5 Pro: 29.2 (#88)

| Benchmark | Claude Sonnet 5 | Gemini 2.5 Pro |
|---|---|---|
| LMArena Search | 1194 | 1142 |
| Vending-Bench 2 | 6,378 | 573.64 |
| Terminal-Bench | — | 32.6% |
| APEX-Agents | 54.5% | — |
| GDPval | — | 23.3% |
| Remote Labor Index | — | 0.8% |
| TheAgentCompany | — | 30.3% |
| τ²-bench Banking | — | 13.7% |
| DeepResearch Bench | — | 42.8% |
| BALROG | — | 43.3% |
| GBAEval | 65.3% | — |
| METR Time Horizons | — | 55.4% |

## Reasoning

- Claude Sonnet 5: 49.1 (#39)
- Gemini 2.5 Pro: 28.8 (#99)

| Benchmark | Claude Sonnet 5 | Gemini 2.5 Pro |
|---|---|---|
| SimpleBench | 60.6% | 62.4% |
| CritPt | 16.9% | 2% |
| Chess Puzzles | 35% | 20% |
| LMArena Hard Prompts | 1461 | 1455 |
| DTBench | 92.5% | 82.4% |
| LMCA | 50% | 34.8% |
| Epoch Capabilities Index | 156.21 | 145.32 |
| ForecastBench | 61.1 | 61.3 |
| ARC-AGI-2 | — | 4.9% |
| Kagi LLM Benchmark | — | 70.3% |
| NYT Connections (extended) | 75.1% | — |
| ARC-AGI-1 | — | 41% |
| EnigmaEval | — | 5.6% |
| LiveBench Reasoning | — | 89.8% |
| Mystery Game Puzzles | 35% | — |
| LiveBench Data Analysis | — | 79.9% |
| Surface Evolver Bench | 60% | — |
| Bench to the Future 3 | 0.14 | — |
| LiveBench | — | 82.3% |

## Math

- Claude Sonnet 5: 66.2 (#27)
- Gemini 2.5 Pro: 32.5 (#213)

| Benchmark | Claude Sonnet 5 | Gemini 2.5 Pro |
|---|---|---|
| FrontierMath (Tiers 1-3) | 65.6% | 24.6% |
| FrontierMath Tier 4 | 29.3% | 0% |
| OTIS Mock AIME 2024-2025 | 94.7% | 84.7% |
| LMArena Math | 1467 | 1450 |
| ProofBench | 77% | — |
| Omni-MATH | — | 41.6% |
| LiveBench Math | — | 90.2% |
| MATH Level 5 | — | 95.9% |
| FrontierMath (Feb 2025 set) | — | 14.1% |
| FrontierMath Tier 4 (v1) | — | 4.2% |

## Knowledge

- Claude Sonnet 5: 55.6 (#47)
- Gemini 2.5 Pro: 56.0 (#46)

| Benchmark | Claude Sonnet 5 | Gemini 2.5 Pro |
|---|---|---|
| GPQA Diamond | 90.5% | 85.3% |
| LMArena Expert | 1490 | 1452 |
| Humanity's Last Exam | — | 21.6% |
| SimpleQA Verified | 33.7% | — |
| MMLU-Pro | — | 86.3% |
| Confabulations | — | 10.6% |
| Vectara Hallucination Rate | — | 7% |
| GPQA (HELM) | — | 74.9% |

## Multimodal

- Claude Sonnet 5: 42.4 (#31)
- Gemini 2.5 Pro: 45.2 (#18)

| Benchmark | Claude Sonnet 5 | Gemini 2.5 Pro |
|---|---|---|
| LMArena Vision | 1274 | 1263 |
| LMArena Document | 1466 | 1421 |
| GeoBench | — | 86% |
| VPCT | — | 48% |
| Blueprint-Bench 2 | 24.9% | — |
| SpatialViz-Bench | — | 44.7% |

## Multilingual

- Claude Sonnet 5: 53.8 (#55)
- Gemini 2.5 Pro: 55.3 (#31)

| Benchmark | Claude Sonnet 5 | Gemini 2.5 Pro |
|---|---|---|
| LMArena Non-English | 1431 | 1451 |
| LMArena Chinese | 1477 | 1507 |
| LMArena French | 1460 | 1472 |
| LMArena German | 1440 | 1487 |
| LMArena Japanese | 1422 | 1461 |
| LMArena Korean | 1411 | 1434 |
| LMArena Russian | 1451 | 1461 |
| LMArena Spanish | 1437 | 1473 |

## Instruction Following

- Claude Sonnet 5: 76.3 (#41)
- Gemini 2.5 Pro: 75.0 (#75)

| Benchmark | Claude Sonnet 5 | Gemini 2.5 Pro |
|---|---|---|
| LMArena Instruction Following | 1452 | 1437 |
| LiveBench Instruction Following | — | 80.6% |
| IFEval | — | 84% |

## Long Context

- Claude Sonnet 5: 44.8 (#55)
- Gemini 2.5 Pro: 59.8 (#5)

| Benchmark | Claude Sonnet 5 | Gemini 2.5 Pro |
|---|---|---|
| LMArena Longer Query | 1463 | 1449 |
| Fiction.LiveBench | — | 91.7% |

## Writing & Preference

- Claude Sonnet 5: 69.2 (#25)
- Gemini 2.5 Pro: 63.7 (#62)

| Benchmark | Claude Sonnet 5 | Gemini 2.5 Pro |
|---|---|---|
| LMArena Text | 1442 | 1458 |
| LMArena Creative Writing | 1416 | 1454 |
| EQ-Bench Creative Writing | 1794 | 1421 |
| LMArena Multi-Turn | 1454 | 1453 |
| Short-Story Creative Writing | — | 83.8% |
| WildBench | — | 85.7% |
| EQ-Bench 4 | 1236 | — |
| LiveBench Language | — | 67.8% |

## FAQ

### Is Claude Sonnet 5 better than Gemini 2.5 Pro?

Claude Sonnet 5 is the stronger model overall, scoring 54.6 to 45.0 on the Noometry Index.

### Which is cheaper, Claude Sonnet 5 or Gemini 2.5 Pro?

Gemini 2.5 Pro is cheaper. It lists at $1.25 per million input tokens and $10 per million output tokens; Claude Sonnet 5 lists at $2 and $10.

### Is Claude Sonnet 5 or Gemini 2.5 Pro better for coding?

Claude Sonnet 5 scores higher on coding benchmarks: 55.5 versus 42.4 in the Noometry coding category.

### Which has the bigger context window?

Gemini 2.5 Pro does, with 1.05M tokens against 1M.

### How many benchmarks do Claude Sonnet 5 and Gemini 2.5 Pro share?

38 benchmarks have published results for both models. Claude Sonnet 5 has 51 scored results on Noometry and Gemini 2.5 Pro has 78.