# Claude 2 vs Claude Sonnet 4

> Claude Sonnet 4 is the stronger model overall, scoring 40.8 to 25.0 on the Noometry Index.

- Canonical page: https://noometry.com/compare/claude-2-vs-claude-sonnet-4
- Last updated: 2026-10-11
- Shared benchmarks: 5

## Summary

- They share 5 benchmarks with published results for both. Claude 2 scores higher in 0 categories and Claude Sonnet 4 in 3 categories; 3 gaps are clear of the uncertainty.
- The widest gap is in math, where Claude Sonnet 4 leads 43.3 to 9.3.
- The biggest single-benchmark swing is MATH Level 5: 11.7% for Claude 2 and 84.4% for Claude Sonnet 4.

## Snapshot

| | Claude 2 | Claude Sonnet 4 |
|---|---|---|
| Provider | Anthropic | Anthropic |
| Noometry Index | 25.0 | 40.8 |
| Rank | 346 | 145 |
| Context | — | 200K |
| Input $/M | — | $3 |
| Output $/M | — | $15 |
| Weights | Proprietary | Proprietary |

## Coding

- Claude 2: —
- Claude Sonnet 4: 43.5 (#88)

| Benchmark | Claude 2 | Claude Sonnet 4 |
|---|---|---|
| SWE-bench Verified (bash only) | — | 64.9% |
| Aider Polyglot | — | 61.3% |
| SciCode | — | 40% |
| GSO | — | 4.9% |
| WeirdML | — | 46.1% |
| LMArena Coding | — | 1414 |
| ALE-Bench | — | 655.35 |
| HumanEval+ | 61.6% | — |

## Agentic & Tool Use

- Claude 2: —
- Claude Sonnet 4: 38.5 (#31)

| Benchmark | Claude 2 | Claude Sonnet 4 |
|---|---|---|
| TheAgentCompany | — | 33.1% |
| Cybench | — | 35% |
| DeepResearch Bench | — | 46.6% |
| OSWorld | — | 43.9% |
| METR Time Horizons | — | 62% |

## Reasoning

- Claude 2: 21.7 (#216)
- Claude Sonnet 4: 22.9 (#187)

| Benchmark | Claude 2 | Claude Sonnet 4 |
|---|---|---|
| DTBench | 51.9% | 77.1% |
| Epoch Capabilities Index | 120.13 | 141.69 |
| ARC-AGI-2 | — | 5.9% |
| SimpleBench | — | 45.5% |
| Kagi LLM Benchmark | — | 73% |
| ARC-AGI-1 | — | 40% |
| CritPt | — | 0.3% |
| EnigmaEval | — | 3.1% |
| LMArena Hard Prompts | — | 1372 |
| LMCA | — | 29% |
| ForecastBench | — | 60.2 |

## Math

- Claude 2: 9.3 (#320)
- Claude Sonnet 4: 43.3 (#80)

| Benchmark | Claude 2 | Claude Sonnet 4 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 2.5% | 71.1% |
| MATH Level 5 | 11.7% | 84.4% |
| Omni-MATH | — | 60.2% |
| LMArena Math | — | 1375 |
| FrontierMath (Feb 2025 set) | — | 4.1% |
| FrontierMath Tier 4 (v1) | — | 0% |

## Knowledge

- Claude 2: 16.9 (#287)
- Claude Sonnet 4: 41.8 (#108)

| Benchmark | Claude 2 | Claude Sonnet 4 |
|---|---|---|
| GPQA Diamond | 34.7% | 79.2% |
| Humanity's Last Exam | — | 7.8% |
| MMLU-Pro | — | 84.3% |
| Confabulations | — | 13.2% |
| Vectara Hallucination Rate | — | 10.3% |
| GPQA (HELM) | — | 70.6% |
| LMArena Expert | — | 1372 |
| MMLU | 78.5% | — |
| TriviaQA | 87.5% | — |

## Multimodal

- Claude 2: —
- Claude Sonnet 4: 26.2 (#121)

| Benchmark | Claude 2 | Claude Sonnet 4 |
|---|---|---|
| LMArena Vision | — | 1191 |
| GeoBench | — | 37% |
| VPCT | — | 34% |
| MindCube | — | 44.8% |

## Multilingual

- Claude 2: —
- Claude Sonnet 4: 46.7 (#156)

| Benchmark | Claude 2 | Claude Sonnet 4 |
|---|---|---|
| LMArena Non-English | — | 1333 |
| LMArena Chinese | — | 1350 |
| LMArena French | — | 1363 |
| LMArena German | — | 1331 |
| LMArena Japanese | — | 1302 |
| LMArena Korean | — | 1291 |
| LMArena Russian | — | 1355 |
| LMArena Spanish | — | 1357 |

## Instruction Following

- Claude 2: —
- Claude Sonnet 4: 71.7 (#145)

| Benchmark | Claude 2 | Claude Sonnet 4 |
|---|---|---|
| IFEval | — | 84% |
| LMArena Instruction Following | — | 1376 |

## Long Context

- Claude 2: —
- Claude Sonnet 4: 33.7 (#259)

| Benchmark | Claude 2 | Claude Sonnet 4 |
|---|---|---|
| Fiction.LiveBench | — | 46.9% |
| LMArena Longer Query | — | 1398 |

## Writing & Preference

- Claude 2: —
- Claude Sonnet 4: 57.1 (#132)

| Benchmark | Claude 2 | Claude Sonnet 4 |
|---|---|---|
| LMArena Text | — | 1351 |
| LMArena Creative Writing | — | 1345 |
| Short-Story Creative Writing | — | 81.4% |
| EQ-Bench Creative Writing | — | 1483 |
| WildBench | — | 83.8% |
| LMArena Multi-Turn | — | 1376 |

## FAQ

### Is Claude 2 better than Claude Sonnet 4?

Claude Sonnet 4 is the stronger model overall, scoring 40.8 to 25.0 on the Noometry Index.

### How many benchmarks do Claude 2 and Claude Sonnet 4 share?

5 benchmarks have published results for both models. Claude 2 has 8 scored results on Noometry and Claude Sonnet 4 has 58.