# Claude Sonnet 4 vs Gemini 3 Flash Preview

> Gemini 3 Flash Preview is the stronger model overall, scoring 52.3 to 40.6 on the Noometry Index.

- Canonical page: https://noometry.com/compare/claude-sonnet-4-vs-gemini-3-flash-preview
- Last updated: 2026-10-10
- Shared benchmarks: 37

## Summary

- They share 37 benchmarks with published results for both. Claude Sonnet 4 scores higher in 0 categories and Gemini 3 Flash Preview in 10 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Gemini 3 Flash Preview leads 49.2 to 23.0.
- The biggest single-benchmark swing is GeoBench: 37% for Claude Sonnet 4 and 88% for Gemini 3 Flash Preview.
- Gemini 3 Flash Preview is cheaper at $0.50 / $3 per million input/output tokens, against $3 / $15 for Claude Sonnet 4.
- Gemini 3 Flash Preview accepts more context: 1.05M tokens versus 200K.

## Snapshot

| | Claude Sonnet 4 | Gemini 3 Flash Preview |
|---|---|---|
| Provider | Anthropic | Google |
| Noometry Index | 40.6 | 52.3 |
| Rank | 139 | 33 |
| Context | 200K | 1.05M |
| Input $/M | $3 | $0.50 |
| Output $/M | $15 | $3 |
| Weights | Proprietary | Proprietary |

## Coding

- Claude Sonnet 4: 44.0 (#70)
- Gemini 3 Flash Preview: 51.5 (#35)

| Benchmark | Claude Sonnet 4 | Gemini 3 Flash Preview |
|---|---|---|
| SWE-bench Verified (bash only) | 64.9% | 75.8% |
| GSO | 4.9% | 9.8% |
| WeirdML | 46.1% | 61.6% |
| LMArena Coding | 1414 | 1460 |
| ALE-Bench | 655.35 | 1,367 |
| SWE-bench Verified | — | 75.4% |
| Aider Polyglot | 61.3% | — |
| LMArena WebDev | — | 1439 |
| SWE-bench Multilingual | — | 72.7% |
| SciCode | 40% | — |

## Agentic & Tool Use

- Claude Sonnet 4: 38.3 (#29)
- Gemini 3 Flash Preview: 38.6 (#27)

| Benchmark | Claude Sonnet 4 | Gemini 3 Flash Preview |
|---|---|---|
| DeepResearch Bench | 46.6% | 49.8% |
| Terminal-Bench | — | 64.3% |
| TheAgentCompany | 33.1% | — |
| τ²-bench Airline | — | 82.5% |
| τ²-bench Banking | — | 27.3% |
| τ²-bench Retail | — | 76.8% |
| τ²-bench Telecom | — | 91.2% |
| Cybench | 35% | — |
| OSWorld | 43.9% | — |
| BALROG | — | 48.1% |
| GDP.pdf | — | 10% |
| LMArena Search | — | 1198 |
| METR Time Horizons | 62% | — |
| Vending-Bench 2 | — | 3,635 |

## Reasoning

- Claude Sonnet 4: 23.0 (#169)
- Gemini 3 Flash Preview: 49.2 (#35)

| Benchmark | Claude Sonnet 4 | Gemini 3 Flash Preview |
|---|---|---|
| ARC-AGI-2 | 5.9% | 33.6% |
| SimpleBench | 45.5% | 61.1% |
| ARC-AGI-1 | 40% | 84.7% |
| LMArena Hard Prompts | 1372 | 1465 |
| DTBench | 77.1% | 89.1% |
| LMCA | 29% | 43.1% |
| Epoch Capabilities Index | 141.69 | 151.83 |
| ForecastBench | 60.2 | 58.5 |
| Kagi LLM Benchmark | 73% | — |
| NYT Connections (extended) | — | 83.1% |
| CritPt | 0.3% | — |
| Chess Puzzles | — | 40% |
| EnigmaEval | 3.1% | — |
| Mystery Game Puzzles | — | 26% |

## Math

- Claude Sonnet 4: 41.5 (#71)
- Gemini 3 Flash Preview: 51.4 (#42)

| Benchmark | Claude Sonnet 4 | Gemini 3 Flash Preview |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 71.1% | 95.6% |
| LMArena Math | 1375 | 1472 |
| FrontierMath (Feb 2025 set) | 4.1% | 35.6% |
| FrontierMath Tier 4 (v1) | 0% | 4.2% |
| FrontierMath (Tiers 1-3) | — | 51.2% |
| FrontierMath Tier 4 | — | 17.1% |
| MathArena Final-Answer Competitions | — | 67.6% |
| ProofBench | — | 15% |
| Omni-MATH | 60.2% | — |
| MATH Level 5 | 84.4% | — |

## Knowledge

- Claude Sonnet 4: 41.6 (#99)
- Gemini 3 Flash Preview: 58.8 (#30)

| Benchmark | Claude Sonnet 4 | Gemini 3 Flash Preview |
|---|---|---|
| GPQA Diamond | 79.2% | 89.4% |
| Vectara Hallucination Rate | 10.3% | 13.5% |
| LMArena Expert | 1371 | 1461 |
| Humanity's Last Exam | 7.8% | — |
| SimpleQA Verified | — | 66.8% |
| MMLU-Pro | 84.3% | — |
| Confabulations | 13.2% | — |
| GPQA (HELM) | 70.6% | — |

## Multimodal

- Claude Sonnet 4: 25.6 (#116)
- Gemini 3 Flash Preview: 45.3 (#15)

| Benchmark | Claude Sonnet 4 | Gemini 3 Flash Preview |
|---|---|---|
| LMArena Vision | 1191 | 1285 |
| GeoBench | 37% | 88% |
| VPCT | 34% | 72.6% |
| Blueprint-Bench 2 | — | 0% |
| LMArena Document | — | 1413 |
| MindCube | 44.8% | — |

## Multilingual

- Claude Sonnet 4: 46.9 (#151)
- Gemini 3 Flash Preview: 55.9 (#26)

| Benchmark | Claude Sonnet 4 | Gemini 3 Flash Preview |
|---|---|---|
| LMArena Non-English | 1333 | 1458 |
| LMArena Chinese | 1351 | 1510 |
| LMArena French | 1364 | 1478 |
| LMArena German | 1331 | 1497 |
| LMArena Japanese | 1301 | 1488 |
| LMArena Korean | 1293 | 1443 |
| LMArena Russian | 1355 | 1480 |
| LMArena Spanish | 1356 | 1470 |

## Instruction Following

- Claude Sonnet 4: 71.8 (#141)
- Gemini 3 Flash Preview: 75.8 (#52)

| Benchmark | Claude Sonnet 4 | Gemini 3 Flash Preview |
|---|---|---|
| LMArena Instruction Following | 1375 | 1437 |
| IFEval | 84% | — |

## Long Context

- Claude Sonnet 4: 33.6 (#254)
- Gemini 3 Flash Preview: 44.4 (#63)

| Benchmark | Claude Sonnet 4 | Gemini 3 Flash Preview |
|---|---|---|
| LMArena Longer Query | 1398 | 1452 |
| Fiction.LiveBench | 46.9% | — |

## Writing & Preference

- Claude Sonnet 4: 57.3 (#127)
- Gemini 3 Flash Preview: 65.6 (#44)

| Benchmark | Claude Sonnet 4 | Gemini 3 Flash Preview |
|---|---|---|
| LMArena Text | 1351 | 1466 |
| LMArena Creative Writing | 1345 | 1457 |
| LMArena Multi-Turn | 1376 | 1471 |
| Short-Story Creative Writing | 81.4% | — |
| EQ-Bench Creative Writing | 1483 | — |
| WildBench | 83.8% | — |

## FAQ

### Is Claude Sonnet 4 better than Gemini 3 Flash Preview?

Gemini 3 Flash Preview is the stronger model overall, scoring 52.3 to 40.6 on the Noometry Index.

### Which is cheaper, Claude Sonnet 4 or Gemini 3 Flash Preview?

Gemini 3 Flash Preview is cheaper. It lists at $0.50 per million input tokens and $3 per million output tokens; Claude Sonnet 4 lists at $3 and $15.

### Is Claude Sonnet 4 or Gemini 3 Flash Preview better for coding?

Gemini 3 Flash Preview scores higher on coding benchmarks: 51.5 versus 44.0 in the Noometry coding category.

### Which has the bigger context window?

Gemini 3 Flash Preview does, with 1.05M tokens against 200K.

### How many benchmarks do Claude Sonnet 4 and Gemini 3 Flash Preview share?

37 benchmarks have published results for both models. Claude Sonnet 4 has 58 scored results on Noometry and Gemini 3 Flash Preview has 59.