# DeepSeek-V3.1 vs Qwen-14B

> DeepSeek-V3.1 is the stronger model overall, scoring 42.2 to 30.8 on the Noometry Index.

- Canonical page: https://noometry.com/compare/deepseek-v3-1-vs-qwen-14b
- Last updated: 2026-10-10
- Shared benchmarks: 11

## Summary

- They share 11 benchmarks with published results for both. DeepSeek-V3.1 scores higher in 7 categories and Qwen-14B in 0 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where DeepSeek-V3.1 leads 60.5 to 27.8.

## Snapshot

| | DeepSeek-V3.1 | Qwen-14B |
|---|---|---|
| Provider | DeepSeek | Alibaba (Qwen) |
| Noometry Index | 42.2 | 30.8 |
| Rank | 106 | 273 |
| Context | 164K | — |
| Input $/M | $0.25 | — |
| Output $/M | $0.95 | — |
| Weights | Open | Open |

## Coding

- DeepSeek-V3.1: 39.8 (#140)
- Qwen-14B: 30.6 (#283)

| Benchmark | DeepSeek-V3.1 | Qwen-14B |
|---|---|---|
| LMArena Coding | 1417 | 1072 |
| WeirdML | 38.4% | — |

## Reasoning

- DeepSeek-V3.1: 27.3 (#109)
- Qwen-14B: 18.7 (#255)

| Benchmark | DeepSeek-V3.1 | Qwen-14B |
|---|---|---|
| LMArena Hard Prompts | 1417 | 1027 |
| Epoch Capabilities Index | 139.92 | 112.8 |
| SimpleBench | 40% | — |
| Kagi LLM Benchmark | 53.2% | — |
| DTBench | 82.7% | — |
| LMCA | 24.3% | — |
| BIG-Bench Hard | — | 55% |
| ForecastBench | 58 | — |
| LAMBADA | — | 71.1% |
| PIQA | — | 79.9% |

## Math

- DeepSeek-V3.1: 36.1 (#125)
- Qwen-14B: 28.7 (#224)

| Benchmark | DeepSeek-V3.1 | Qwen-14B |
|---|---|---|
| LMArena Math | 1420 | 1068 |
| GSM8K | — | 61.3% |

## Knowledge

- DeepSeek-V3.1: 43.0 (#86)
- Qwen-14B: —

| Benchmark | DeepSeek-V3.1 | Qwen-14B |
|---|---|---|
| Vectara Hallucination Rate | 5.5% | — |
| LMArena Expert | 1403 | — |
| ARC (AI2) Challenge | — | 84.4% |
| BoolQ | — | 86.2% |
| MMLU | — | 66.3% |

## Multilingual

- DeepSeek-V3.1: 51.8 (#100)
- Qwen-14B: 27.7 (#270)

| Benchmark | DeepSeek-V3.1 | Qwen-14B |
|---|---|---|
| LMArena Non-English | 1400 | 1042 |
| LMArena Chinese | 1470 | 1077 |
| LMArena French | 1447 | — |
| LMArena German | 1412 | — |
| LMArena Japanese | 1374 | — |
| LMArena Korean | 1337 | — |
| LMArena Russian | 1405 | — |
| LMArena Spanish | 1429 | — |

## Instruction Following

- DeepSeek-V3.1: 74.1 (#105)
- Qwen-14B: 52.6 (#284)

| Benchmark | DeepSeek-V3.1 | Qwen-14B |
|---|---|---|
| LMArena Instruction Following | 1400 | 1031 |

## Long Context

- DeepSeek-V3.1: 36.3 (#227)
- Qwen-14B: 31.3 (#276)

| Benchmark | DeepSeek-V3.1 | Qwen-14B |
|---|---|---|
| LMArena Longer Query | 1422 | 1029 |
| Fiction.LiveBench | 52.8% | — |

## Writing & Preference

- DeepSeek-V3.1: 60.5 (#93)
- Qwen-14B: 27.8 (#294)

| Benchmark | DeepSeek-V3.1 | Qwen-14B |
|---|---|---|
| LMArena Text | 1420 | 1051 |
| LMArena Creative Writing | 1401 | 1028 |
| LMArena Multi-Turn | 1408 | 1022 |
| EQ-Bench Creative Writing | 1436 | — |

## FAQ

### Is DeepSeek-V3.1 better than Qwen-14B?

DeepSeek-V3.1 is the stronger model overall, scoring 42.2 to 30.8 on the Noometry Index.

### Is DeepSeek-V3.1 or Qwen-14B better for coding?

DeepSeek-V3.1 scores higher on coding benchmarks: 39.8 versus 30.6 in the Noometry coding category.

### How many benchmarks do DeepSeek-V3.1 and Qwen-14B share?

11 benchmarks have published results for both models. DeepSeek-V3.1 has 27 scored results on Noometry and Qwen-14B has 18.