# Claude 3 Sonnet vs DeepSeek-V3.2-Exp

> DeepSeek-V3.2-Exp is the stronger model overall, scoring 44.9 to 28.8 on the Noometry Index.

- Canonical page: https://noometry.com/compare/claude-3-sonnet-vs-deepseek-v3-2-exp
- Last updated: 2026-10-10
- Shared benchmarks: 22

## Summary

- They share 22 benchmarks with published results for both. Claude 3 Sonnet scores higher in 0 categories and DeepSeek-V3.2-Exp in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where DeepSeek-V3.2-Exp leads 40.8 to 10.0.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 2.5% for Claude 3 Sonnet and 87.8% for DeepSeek-V3.2-Exp.
- DeepSeek-V3.2-Exp has downloadable open weights; the other is API-only.

## Snapshot

| | Claude 3 Sonnet | DeepSeek-V3.2-Exp |
|---|---|---|
| Provider | Anthropic | DeepSeek |
| Noometry Index | 28.8 | 44.9 |
| Rank | 311 | 68 |
| Context | — | 164K |
| Input $/M | — | $0.26 |
| Output $/M | — | $0.38 |
| Weights | Proprietary | Open |

## Coding

- Claude 3 Sonnet: 29.2 (#296)
- DeepSeek-V3.2-Exp: 46.4 (#61)

| Benchmark | Claude 3 Sonnet | DeepSeek-V3.2-Exp |
|---|---|---|
| WeirdML | 10.2% | 39.5% |
| LMArena Coding | 1223 | 1454 |
| SWE-bench Verified (bash only) | — | 70% |
| Aider Polyglot | — | 74.2% |
| LMArena WebDev | — | 1362 |
| SWE-bench Multilingual | — | 59% |
| SciCode | — | 38.9% |
| BigCodeBench Instruct | 42.7% | — |
| BigCodeBench Complete | 53.8% | — |
| HumanEval+ | 64% | — |
| MBPP+ | 69.3% | — |

## Agentic & Tool Use

- Claude 3 Sonnet: —
- DeepSeek-V3.2-Exp: 37.9 (#31)

| Benchmark | Claude 3 Sonnet | DeepSeek-V3.2-Exp |
|---|---|---|
| Terminal-Bench | — | 39.6% |
| Berkeley Function Calling Leaderboard | — | 56.7% |
| TheAgentCompany | — | 42.9% |
| Vending-Bench 2 | — | 1,034 |

## Reasoning

- Claude 3 Sonnet: 19.7 (#232)
- DeepSeek-V3.2-Exp: 22.0 (#185)

| Benchmark | Claude 3 Sonnet | DeepSeek-V3.2-Exp |
|---|---|---|
| LMArena Hard Prompts | 1197 | 1433 |
| DTBench | 53.6% | 87.7% |
| Epoch Capabilities Index | 120.69 | 146.28 |
| ARC-AGI-2 | — | 4% |
| Kagi LLM Benchmark | — | 52.2% |
| NYT Connections (extended) | — | 36.7% |
| ARC-AGI-1 | — | 57% |
| CritPt | — | 2.9% |
| Chess Puzzles | — | 14% |
| Thematic Generalization | — | 65% |
| LMCA | — | 29.1% |
| WinoGrande | 75.1% | — |

## Math

- Claude 3 Sonnet: 10.0 (#305)
- DeepSeek-V3.2-Exp: 40.8 (#75)

| Benchmark | Claude 3 Sonnet | DeepSeek-V3.2-Exp |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 2.5% | 87.8% |
| LMArena Math | 1213 | 1435 |
| MathArena Final-Answer Competitions | — | 57.7% |
| ProofBench | — | 8% |
| MATH Level 5 | 18.2% | — |
| FrontierMath (Feb 2025 set) | — | 22.1% |
| FrontierMath Tier 4 (v1) | — | 2.1% |

## Knowledge

- Claude 3 Sonnet: 21.1 (#271)
- DeepSeek-V3.2-Exp: 51.6 (#61)

| Benchmark | Claude 3 Sonnet | DeepSeek-V3.2-Exp |
|---|---|---|
| GPQA Diamond | 40.6% | 83.4% |
| LMArena Expert | 1173 | 1436 |
| Vectara Hallucination Rate | — | 5.3% |
| MMLU | 75.9% | — |

## Multimodal

- Claude 3 Sonnet: 24.7 (#120)
- DeepSeek-V3.2-Exp: —

| Benchmark | Claude 3 Sonnet | DeepSeek-V3.2-Exp |
|---|---|---|
| LMArena Vision | 984 | — |

## Multilingual

- Claude 3 Sonnet: 38.0 (#229)
- DeepSeek-V3.2-Exp: 52.4 (#85)

| Benchmark | Claude 3 Sonnet | DeepSeek-V3.2-Exp |
|---|---|---|
| LMArena Non-English | 1205 | 1409 |
| LMArena Chinese | 1188 | 1461 |
| LMArena French | 1228 | 1435 |
| LMArena German | 1204 | 1440 |
| LMArena Japanese | 1131 | 1377 |
| LMArena Korean | 1128 | 1372 |
| LMArena Russian | 1227 | 1423 |
| LMArena Spanish | 1204 | 1441 |

## Instruction Following

- Claude 3 Sonnet: 63.1 (#230)
- DeepSeek-V3.2-Exp: 74.7 (#89)

| Benchmark | Claude 3 Sonnet | DeepSeek-V3.2-Exp |
|---|---|---|
| LMArena Instruction Following | 1200 | 1413 |

## Long Context

- Claude 3 Sonnet: 36.7 (#223)
- DeepSeek-V3.2-Exp: 47.6 (#15)

| Benchmark | Claude 3 Sonnet | DeepSeek-V3.2-Exp |
|---|---|---|
| LMArena Longer Query | 1211 | 1428 |
| Fiction.LiveBench | — | 83.3% |
| CL-bench | — | 13.2% |
| CL-bench Life | — | 9.5% |

## Writing & Preference

- Claude 3 Sonnet: 42.3 (#233)
- DeepSeek-V3.2-Exp: 62.5 (#73)

| Benchmark | Claude 3 Sonnet | DeepSeek-V3.2-Exp |
|---|---|---|
| LMArena Text | 1218 | 1425 |
| LMArena Creative Writing | 1186 | 1403 |
| LMArena Multi-Turn | 1227 | 1427 |
| EQ-Bench Creative Writing | — | 1515 |

## FAQ

### Is Claude 3 Sonnet better than DeepSeek-V3.2-Exp?

DeepSeek-V3.2-Exp is the stronger model overall, scoring 44.9 to 28.8 on the Noometry Index.

### Is Claude 3 Sonnet or DeepSeek-V3.2-Exp better for coding?

DeepSeek-V3.2-Exp scores higher on coding benchmarks: 46.4 versus 29.2 in the Noometry coding category.

### How many benchmarks do Claude 3 Sonnet and DeepSeek-V3.2-Exp share?

22 benchmarks have published results for both models. Claude 3 Sonnet has 30 scored results on Noometry and DeepSeek-V3.2-Exp has 48.