# Claude 3 Haiku vs DeepSeek-V3.1

> DeepSeek-V3.1 is the stronger model overall, scoring 42.8 to 25.9 on the Noometry Index.

- Canonical page: https://noometry.com/compare/claude-3-haiku-vs-deepseek-v3-1
- Last updated: 2026-10-12
- Shared benchmarks: 24

## Summary

- They share 24 benchmarks with published results for both. Claude 3 Haiku scores higher in 0 categories and DeepSeek-V3.1 in 8 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where DeepSeek-V3.1 leads 60.3 to 29.7.
- The biggest single-benchmark swing is DTBench: 50.1% for Claude 3 Haiku and 82.7% for DeepSeek-V3.1.
- DeepSeek-V3.1 has downloadable open weights; the other is API-only.

## Snapshot

| | Claude 3 Haiku | DeepSeek-V3.1 |
|---|---|---|
| Provider | Anthropic | DeepSeek |
| Noometry Index | 25.9 | 42.8 |
| Rank | 340 | 108 |
| Context | — | 164K |
| Input $/M | — | $0.25 |
| Output $/M | — | $0.95 |
| Weights | Proprietary | Open |

## Coding

- Claude 3 Haiku: 26.4 (#325)
- DeepSeek-V3.1: 40.3 (#144)

| Benchmark | Claude 3 Haiku | DeepSeek-V3.1 |
|---|---|---|
| WeirdML | 9.8% | 38.4% |
| LMArena Coding | 1199 | 1417 |
| BigCodeBench Instruct | 39.4% | — |
| BigCodeBench Complete | 50.1% | — |
| CadEval | 12% | — |
| HumanEval+ | 68.9% | — |
| MBPP+ | 68.8% | — |

## Reasoning

- Claude 3 Haiku: 16.3 (#307)
- DeepSeek-V3.1: 27.9 (#110)

| Benchmark | Claude 3 Haiku | DeepSeek-V3.1 |
|---|---|---|
| Kagi LLM Benchmark | 34.2% | 53.2% |
| LMArena Hard Prompts | 1174 | 1417 |
| DTBench | 50.1% | 82.7% |
| LMCA | 8.8% | 24.3% |
| Epoch Capabilities Index | 118.35 | 139.92 |
| ForecastBench | 53.2 | 58 |
| SimpleBench | — | 40% |
| WinoGrande | 74.2% | — |

## Math

- Claude 3 Haiku: 9.8 (#319)
- DeepSeek-V3.1: 38.9 (#122)

| Benchmark | Claude 3 Haiku | DeepSeek-V3.1 |
|---|---|---|
| LMArena Math | 1188 | 1420 |
| OTIS Mock AIME 2024-2025 | 1.8% | — |
| MATH Level 5 | 14.9% | — |

## Knowledge

- Claude 3 Haiku: 17.3 (#285)
- DeepSeek-V3.1: 43.7 (#90)

| Benchmark | Claude 3 Haiku | DeepSeek-V3.1 |
|---|---|---|
| LMArena Expert | 1148 | 1405 |
| GPQA Diamond | 36.3% | — |
| Confabulations | 34.2% | — |
| Vectara Hallucination Rate | — | 5.5% |
| MMLU | 73.8% | — |

## Multimodal

- Claude 3 Haiku: 23.6 (#128)
- DeepSeek-V3.1: —

| Benchmark | Claude 3 Haiku | DeepSeek-V3.1 |
|---|---|---|
| LMArena Vision | 950 | — |
| ScienceQA | 72% | — |

## Multilingual

- Claude 3 Haiku: 36.0 (#243)
- DeepSeek-V3.1: 51.6 (#106)

| Benchmark | Claude 3 Haiku | DeepSeek-V3.1 |
|---|---|---|
| LMArena Non-English | 1178 | 1400 |
| LMArena Chinese | 1155 | 1469 |
| LMArena French | 1195 | 1447 |
| LMArena German | 1174 | 1411 |
| LMArena Japanese | 1102 | 1378 |
| LMArena Korean | 1109 | 1337 |
| LMArena Russian | 1204 | 1405 |
| LMArena Spanish | 1166 | 1431 |

## Instruction Following

- Claude 3 Haiku: 61.3 (#247)
- DeepSeek-V3.1: 73.9 (#110)

| Benchmark | Claude 3 Haiku | DeepSeek-V3.1 |
|---|---|---|
| LMArena Instruction Following | 1173 | 1400 |

## Long Context

- Claude 3 Haiku: 36.1 (#237)
- DeepSeek-V3.1: 36.3 (#232)

| Benchmark | Claude 3 Haiku | DeepSeek-V3.1 |
|---|---|---|
| LMArena Longer Query | 1190 | 1422 |
| Fiction.LiveBench | — | 52.8% |

## Writing & Preference

- Claude 3 Haiku: 29.7 (#291)
- DeepSeek-V3.1: 60.3 (#98)

| Benchmark | Claude 3 Haiku | DeepSeek-V3.1 |
|---|---|---|
| LMArena Text | 1195 | 1420 |
| LMArena Creative Writing | 1157 | 1401 |
| EQ-Bench Creative Writing | 717 | 1436 |
| LMArena Multi-Turn | 1190 | 1408 |

## FAQ

### Is Claude 3 Haiku better than DeepSeek-V3.1?

DeepSeek-V3.1 is the stronger model overall, scoring 42.8 to 25.9 on the Noometry Index.

### Is Claude 3 Haiku or DeepSeek-V3.1 better for coding?

DeepSeek-V3.1 scores higher on coding benchmarks: 40.3 versus 26.4 in the Noometry coding category.

### How many benchmarks do Claude 3 Haiku and DeepSeek-V3.1 share?

24 benchmarks have published results for both models. Claude 3 Haiku has 37 scored results on Noometry and DeepSeek-V3.1 has 27.