# Claude Sonnet 4.5 vs DeepSeek V4.1 Flash

> DeepSeek V4.1 Flash is the stronger model overall, scoring 52.8 to 44.1 on the Noometry Index.

- Canonical page: https://noometry.com/compare/claude-sonnet-4-5-vs-deepseek-v4-1-flash
- Last updated: 2026-10-10
- Shared benchmarks: 32

## Summary

- They share 32 benchmarks with published results for both. Claude Sonnet 4.5 scores higher in 3 categories and DeepSeek V4.1 Flash in 7 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in math, where DeepSeek V4.1 Flash leads 66.7 to 32.3.
- The biggest single-benchmark swing is NYT Connections (extended): 37.3% for Claude Sonnet 4.5 and 89.6% for DeepSeek V4.1 Flash.
- DeepSeek V4.1 Flash is cheaper at $0.15 / $0.60 per million input/output tokens, against $3 / $15 for Claude Sonnet 4.5.
- DeepSeek V4.1 Flash accepts more context: 1M tokens versus 200K.
- DeepSeek V4.1 Flash has downloadable open weights; the other is API-only.

## Snapshot

| | Claude Sonnet 4.5 | DeepSeek V4.1 Flash |
|---|---|---|
| Provider | Anthropic | DeepSeek |
| Noometry Index | 44.1 | 52.8 |
| Rank | 81 | 38 |
| Context | 200K | 1M |
| Input $/M | $3 | $0.15 |
| Output $/M | $15 | $0.60 |
| Weights | Proprietary | Open |

## Coding

- Claude Sonnet 4.5: 47.3 (#61)
- DeepSeek V4.1 Flash: 52.9 (#32)

| Benchmark | Claude Sonnet 4.5 | DeepSeek V4.1 Flash |
|---|---|---|
| LMArena WebDev | 1393 | 1619 |
| SciCode | 44.7% | 51.9% |
| LMArena Coding | 1489 | 1506 |
| ALE-Bench | 796.15 | 1,092 |
| SWE-bench Verified | 71.3% | — |
| SWE-bench Verified (bash only) | 71.4% | — |
| SWE-bench Multilingual | 67% | — |
| GSO | 14.7% | — |
| WeirdML | 47.7% | — |
| AlgoTune | 1.52 | — |

## Agentic & Tool Use

- Claude Sonnet 4.5: 38.3 (#32)
- DeepSeek V4.1 Flash: 31.2 (#69)

| Benchmark | Claude Sonnet 4.5 | DeepSeek V4.1 Flash |
|---|---|---|
| Terminal-Bench | 46.5% | — |
| APEX-Agents | — | 39.5% |
| Berkeley Function Calling Leaderboard | 73.2% | — |
| GDPval | 42.5% | — |
| Remote Labor Index | 2.1% | — |
| τ²-bench Airline | 72% | — |
| τ²-bench Banking | 25.3% | — |
| τ²-bench Retail | 72.4% | — |
| τ²-bench Telecom | 84.9% | — |
| Cybench | 60% | — |
| DeepResearch Bench | 52.6% | — |
| OSWorld | 62.9% | — |
| GDP.pdf | — | 19.8% |
| LMArena Search | 1159 | — |
| METR Time Horizons | 67.4% | — |
| Vending-Bench 2 | 3,839 | — |

## Reasoning

- Claude Sonnet 4.5: 26.9 (#125)
- DeepSeek V4.1 Flash: 50.2 (#36)

| Benchmark | Claude Sonnet 4.5 | DeepSeek V4.1 Flash |
|---|---|---|
| NYT Connections (extended) | 37.3% | 89.6% |
| CritPt | 1.1% | 14.3% |
| LMArena Hard Prompts | 1462 | 1483 |
| Mystery Game Puzzles | 17% | 43% |
| DTBench | 83.2% | 89.9% |
| LMCA | 38.8% | 47% |
| Epoch Capabilities Index | 146.84 | 154.9 |
| ARC-AGI-2 | 13.6% | — |
| SimpleBench | 54.3% | — |
| Kagi LLM Benchmark | 57.9% | — |
| ARC-AGI-1 | 63.7% | — |
| Chess Puzzles | 12% | — |
| EnigmaEval | 6% | — |
| EBR-Bench | 2.4% | — |
| Surface Evolver Bench | — | 46.3% |
| ForecastBench | 61.9 | — |

## Math

- Claude Sonnet 4.5: 32.3 (#216)
- DeepSeek V4.1 Flash: 66.7 (#25)

| Benchmark | Claude Sonnet 4.5 | DeepSeek V4.1 Flash |
|---|---|---|
| FrontierMath (Tiers 1-3) | 23.9% | 67.4% |
| FrontierMath Tier 4 | 2.4% | 26.8% |
| OTIS Mock AIME 2024-2025 | 77.8% | 98.3% |
| ProofBench | 19% | 54% |
| LMArena Math | 1449 | 1477 |
| Omni-MATH | 55.3% | — |
| MATH Level 5 | 97.7% | — |
| FrontierMath (Feb 2025 set) | 15.2% | — |
| FrontierMath Tier 4 (v1) | 4.2% | — |

## Knowledge

- Claude Sonnet 4.5: 48.4 (#76)
- DeepSeek V4.1 Flash: 57.9 (#38)

| Benchmark | Claude Sonnet 4.5 | DeepSeek V4.1 Flash |
|---|---|---|
| GPQA Diamond | 82.3% | 89.8% |
| LMArena Expert | 1482 | 1506 |
| Humanity's Last Exam | 13.7% | — |
| SimpleQA Verified | 30.7% | — |
| MMLU-Pro | 86.9% | — |
| Vectara Hallucination Rate | 12% | — |
| GPQA (HELM) | 68.6% | — |

## Multimodal

- Claude Sonnet 4.5: 34.8 (#89)
- DeepSeek V4.1 Flash: 39.1 (#61)

| Benchmark | Claude Sonnet 4.5 | DeepSeek V4.1 Flash |
|---|---|---|
| LMArena Vision | — | 1277 |
| VPCT | 39.8% | — |
| Furniture Assembly | — | 34.2% |
| LMArena Document | 1450 | — |

## Multilingual

- Claude Sonnet 4.5: 53.4 (#69)
- DeepSeek V4.1 Flash: 55.0 (#35)

| Benchmark | Claude Sonnet 4.5 | DeepSeek V4.1 Flash |
|---|---|---|
| LMArena Non-English | 1425 | 1448 |
| LMArena Chinese | 1459 | 1497 |
| LMArena French | 1458 | 1452 |
| LMArena German | 1427 | 1484 |
| LMArena Japanese | 1390 | 1412 |
| LMArena Korean | 1403 | 1452 |
| LMArena Russian | 1437 | 1471 |
| LMArena Spanish | 1457 | 1459 |

## Instruction Following

- Claude Sonnet 4.5: 75.0 (#78)
- DeepSeek V4.1 Flash: 77.3 (#26)

| Benchmark | Claude Sonnet 4.5 | DeepSeek V4.1 Flash |
|---|---|---|
| LMArena Instruction Following | 1459 | 1474 |
| IFEval | 85% | — |

## Long Context

- Claude Sonnet 4.5: 45.2 (#46)
- DeepSeek V4.1 Flash: 45.2 (#47)

| Benchmark | Claude Sonnet 4.5 | DeepSeek V4.1 Flash |
|---|---|---|
| LMArena Longer Query | 1476 | 1475 |

## Writing & Preference

- Claude Sonnet 4.5: 66.5 (#34)
- DeepSeek V4.1 Flash: 65.4 (#48)

| Benchmark | Claude Sonnet 4.5 | DeepSeek V4.1 Flash |
|---|---|---|
| LMArena Text | 1439 | 1462 |
| LMArena Creative Writing | 1442 | 1435 |
| EQ-Bench Creative Writing | 1678 | 1540 |
| LMArena Multi-Turn | 1465 | 1457 |
| WildBench | 85.4% | — |

## FAQ

### Is Claude Sonnet 4.5 better than DeepSeek V4.1 Flash?

DeepSeek V4.1 Flash is the stronger model overall, scoring 52.8 to 44.1 on the Noometry Index.

### Which is cheaper, Claude Sonnet 4.5 or DeepSeek V4.1 Flash?

DeepSeek V4.1 Flash is cheaper. It lists at $0.15 per million input tokens and $0.60 per million output tokens; Claude Sonnet 4.5 lists at $3 and $15.

### Is Claude Sonnet 4.5 or DeepSeek V4.1 Flash better for coding?

DeepSeek V4.1 Flash scores higher on coding benchmarks: 52.9 versus 47.3 in the Noometry coding category.

### Which has the bigger context window?

DeepSeek V4.1 Flash does, with 1M tokens against 200K.

### How many benchmarks do Claude Sonnet 4.5 and DeepSeek V4.1 Flash share?

32 benchmarks have published results for both models. Claude Sonnet 4.5 has 73 scored results on Noometry and DeepSeek V4.1 Flash has 37.