# Claude Sonnet 4 vs DeepSeek-V3.1-Terminus

> DeepSeek-V3.1-Terminus is the stronger model overall, scoring 43.1 to 40.8 on the Noometry Index.

- Canonical page: https://noometry.com/compare/claude-sonnet-4-vs-deepseek-v3-1-terminus
- Last updated: 2026-10-12
- Shared benchmarks: 16

## Summary

- They share 16 benchmarks with published results for both. Claude Sonnet 4 scores higher in 2 categories and DeepSeek-V3.1-Terminus in 5 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in long context, where DeepSeek-V3.1-Terminus leads 43.4 to 33.7.
- The biggest single-benchmark swing is Kagi LLM Benchmark: 73% for Claude Sonnet 4 and 57.4% for DeepSeek-V3.1-Terminus.
- DeepSeek-V3.1-Terminus is cheaper at $0.27 / $1 per million input/output tokens, against $3 / $15 for Claude Sonnet 4.
- Claude Sonnet 4 accepts more context: 200K tokens versus 164K.
- DeepSeek-V3.1-Terminus has downloadable open weights; the other is API-only.

## Snapshot

| | Claude Sonnet 4 | DeepSeek-V3.1-Terminus |
|---|---|---|
| Provider | Anthropic | DeepSeek |
| Noometry Index | 40.8 | 43.1 |
| Rank | 145 | 97 |
| Context | 200K | 164K |
| Input $/M | $3 | $0.27 |
| Output $/M | $15 | $1 |
| Weights | Proprietary | Open |

## Coding

- Claude Sonnet 4: 43.5 (#88)
- DeepSeek-V3.1-Terminus: 42.0 (#113)

| Benchmark | Claude Sonnet 4 | DeepSeek-V3.1-Terminus |
|---|---|---|
| SciCode | 40% | 40.6% |
| LMArena Coding | 1414 | 1426 |
| ALE-Bench | 655.35 | 745.17 |
| SWE-bench Verified (bash only) | 64.9% | — |
| Aider Polyglot | 61.3% | — |
| GSO | 4.9% | — |
| WeirdML | 46.1% | — |

## Agentic & Tool Use

- Claude Sonnet 4: 38.5 (#31)
- DeepSeek-V3.1-Terminus: —

| Benchmark | Claude Sonnet 4 | DeepSeek-V3.1-Terminus |
|---|---|---|
| TheAgentCompany | 33.1% | — |
| Cybench | 35% | — |
| DeepResearch Bench | 46.6% | — |
| OSWorld | 43.9% | — |
| METR Time Horizons | 62% | — |

## Reasoning

- Claude Sonnet 4: 22.9 (#187)
- DeepSeek-V3.1-Terminus: 26.4 (#133)

| Benchmark | Claude Sonnet 4 | DeepSeek-V3.1-Terminus |
|---|---|---|
| Kagi LLM Benchmark | 73% | 57.4% |
| CritPt | 0.3% | 1.7% |
| LMArena Hard Prompts | 1372 | 1426 |
| DTBench | 77.1% | 81.3% |
| LMCA | 29% | 28.6% |
| ARC-AGI-2 | 5.9% | — |
| SimpleBench | 45.5% | — |
| ARC-AGI-1 | 40% | — |
| EnigmaEval | 3.1% | — |
| Epoch Capabilities Index | 141.69 | — |
| ForecastBench | 60.2 | — |

## Math

- Claude Sonnet 4: 43.3 (#80)
- DeepSeek-V3.1-Terminus: 38.5 (#137)

| Benchmark | Claude Sonnet 4 | DeepSeek-V3.1-Terminus |
|---|---|---|
| LMArena Math | 1375 | 1402 |
| OTIS Mock AIME 2024-2025 | 71.1% | — |
| Omni-MATH | 60.2% | — |
| MATH Level 5 | 84.4% | — |
| FrontierMath (Feb 2025 set) | 4.1% | — |
| FrontierMath Tier 4 (v1) | 0% | — |

## Knowledge

- Claude Sonnet 4: 41.8 (#108)
- DeepSeek-V3.1-Terminus: —

| Benchmark | Claude Sonnet 4 | DeepSeek-V3.1-Terminus |
|---|---|---|
| GPQA Diamond | 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 | — |

## Multimodal

- Claude Sonnet 4: 26.2 (#121)
- DeepSeek-V3.1-Terminus: —

| Benchmark | Claude Sonnet 4 | DeepSeek-V3.1-Terminus |
|---|---|---|
| LMArena Vision | 1191 | — |
| GeoBench | 37% | — |
| VPCT | 34% | — |
| MindCube | 44.8% | — |

## Multilingual

- Claude Sonnet 4: 46.7 (#156)
- DeepSeek-V3.1-Terminus: 52.1 (#92)

| Benchmark | Claude Sonnet 4 | DeepSeek-V3.1-Terminus |
|---|---|---|
| LMArena Non-English | 1333 | 1407 |
| LMArena Russian | 1355 | 1436 |
| LMArena Chinese | 1350 | — |
| LMArena French | 1363 | — |
| LMArena German | 1331 | — |
| LMArena Japanese | 1302 | — |
| LMArena Korean | 1291 | — |
| LMArena Spanish | 1357 | — |

## Instruction Following

- Claude Sonnet 4: 71.7 (#145)
- DeepSeek-V3.1-Terminus: 74.0 (#106)

| Benchmark | Claude Sonnet 4 | DeepSeek-V3.1-Terminus |
|---|---|---|
| LMArena Instruction Following | 1376 | 1404 |
| IFEval | 84% | — |

## Long Context

- Claude Sonnet 4: 33.7 (#259)
- DeepSeek-V3.1-Terminus: 43.4 (#97)

| Benchmark | Claude Sonnet 4 | DeepSeek-V3.1-Terminus |
|---|---|---|
| LMArena Longer Query | 1398 | 1421 |
| Fiction.LiveBench | 46.9% | — |

## Writing & Preference

- Claude Sonnet 4: 57.1 (#132)
- DeepSeek-V3.1-Terminus: 61.0 (#92)

| Benchmark | Claude Sonnet 4 | DeepSeek-V3.1-Terminus |
|---|---|---|
| LMArena Text | 1351 | 1419 |
| LMArena Creative Writing | 1345 | 1403 |
| LMArena Multi-Turn | 1376 | 1411 |
| Short-Story Creative Writing | 81.4% | — |
| EQ-Bench Creative Writing | 1483 | — |
| WildBench | 83.8% | — |

## FAQ

### Is Claude Sonnet 4 better than DeepSeek-V3.1-Terminus?

DeepSeek-V3.1-Terminus is the stronger model overall, scoring 43.1 to 40.8 on the Noometry Index.

### Which is cheaper, Claude Sonnet 4 or DeepSeek-V3.1-Terminus?

DeepSeek-V3.1-Terminus is cheaper. It lists at $0.27 per million input tokens and $1 per million output tokens; Claude Sonnet 4 lists at $3 and $15.

### Is Claude Sonnet 4 or DeepSeek-V3.1-Terminus better for coding?

Claude Sonnet 4 scores higher on coding benchmarks: 43.5 versus 42.0 in the Noometry coding category.

### Which has the bigger context window?

Claude Sonnet 4 does, with 200K tokens against 164K.

### How many benchmarks do Claude Sonnet 4 and DeepSeek-V3.1-Terminus share?

16 benchmarks have published results for both models. Claude Sonnet 4 has 58 scored results on Noometry and DeepSeek-V3.1-Terminus has 16.