# Deepseek Coder v2 vs Qwen3 32B

> Qwen3 32B is the stronger model overall, scoring 39.2 to 35.9 on the Noometry Index.

- Canonical page: https://noometry.com/compare/deepseek-coder-v2-vs-qwen3-32b
- Last updated: 2026-10-11
- Shared benchmarks: 13

## Summary

- They share 13 benchmarks with published results for both. Deepseek Coder v2 scores higher in 2 categories and Qwen3 32B in 6 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where Qwen3 32B leads 52.9 to 38.2.

## Snapshot

| | Deepseek Coder v2 | Qwen3 32B |
|---|---|---|
| Provider | DeepSeek | Alibaba (Qwen) |
| Noometry Index | 35.9 | 39.2 |
| Rank | 220 | 172 |
| Context | — | 131K |
| Input $/M | — | $0.70 |
| Output $/M | — | $2.80 |
| Weights | Open | Open |

## Coding

- Deepseek Coder v2: 38.1 (#183)
- Qwen3 32B: 37.7 (#190)

| Benchmark | Deepseek Coder v2 | Qwen3 32B |
|---|---|---|
| LMArena Coding | 1251 | 1358 |
| Aider Polyglot | — | 40% |
| SciCode | — | 35.4% |
| BigCodeBench Instruct | 48.2% | — |
| BigCodeBench Complete | 59.7% | — |
| HumanEval+ | 82.3% | — |
| MBPP+ | 75.1% | — |

## Agentic & Tool Use

- Deepseek Coder v2: —
- Qwen3 32B: 32.6 (#62)

| Benchmark | Deepseek Coder v2 | Qwen3 32B |
|---|---|---|
| Berkeley Function Calling Leaderboard | — | 48.7% |

## Reasoning

- Deepseek Coder v2: 23.6 (#176)
- Qwen3 32B: 20.2 (#241)

| Benchmark | Deepseek Coder v2 | Qwen3 32B |
|---|---|---|
| LMArena Hard Prompts | 1207 | 1334 |
| Kagi LLM Benchmark | — | 54.9% |
| CritPt | — | 0.3% |
| Chess Puzzles | — | 5% |
| DTBench | — | 67.5% |
| LMCA | — | 17.3% |
| Epoch Capabilities Index | — | 138.51 |
| WinoGrande | 83.7% | — |

## Math

- Deepseek Coder v2: 34.9 (#190)
- Qwen3 32B: 39.7 (#99)

| Benchmark | Deepseek Coder v2 | Qwen3 32B |
|---|---|---|
| LMArena Math | 1241 | 1399 |
| OTIS Mock AIME 2024-2025 | — | 66.9% |
| GSM8K | 94.5% | — |

## Knowledge

- Deepseek Coder v2: 32.3 (#212)
- Qwen3 32B: 40.0 (#125)

| Benchmark | Deepseek Coder v2 | Qwen3 32B |
|---|---|---|
| LMArena Expert | 1181 | 1362 |
| GPQA Diamond | — | 65.7% |
| Vectara Hallucination Rate | — | 5.9% |
| ARC (AI2) Challenge | 64.3% | — |

## Multilingual

- Deepseek Coder v2: 36.3 (#240)
- Qwen3 32B: 45.6 (#167)

| Benchmark | Deepseek Coder v2 | Qwen3 32B |
|---|---|---|
| LMArena Non-English | 1182 | 1317 |
| LMArena Chinese | 1201 | 1357 |
| LMArena German | 1164 | 1341 |
| LMArena Russian | 1188 | 1311 |
| LMArena French | 1185 | — |
| LMArena Japanese | 1126 | — |
| LMArena Korean | 1104 | — |
| LMArena Spanish | 1153 | — |

## Instruction Following

- Deepseek Coder v2: 61.7 (#242)
- Qwen3 32B: 68.9 (#179)

| Benchmark | Deepseek Coder v2 | Qwen3 32B |
|---|---|---|
| LMArena Instruction Following | 1180 | 1305 |

## Long Context

- Deepseek Coder v2: 37.0 (#224)
- Qwen3 32B: 43.8 (#87)

| Benchmark | Deepseek Coder v2 | Qwen3 32B |
|---|---|---|
| LMArena Longer Query | 1219 | 1327 |
| Fiction.LiveBench | — | 74.2% |

## Writing & Preference

- Deepseek Coder v2: 38.2 (#253)
- Qwen3 32B: 52.9 (#163)

| Benchmark | Deepseek Coder v2 | Qwen3 32B |
|---|---|---|
| LMArena Text | 1191 | 1340 |
| LMArena Creative Writing | 1120 | 1297 |
| LMArena Multi-Turn | 1177 | 1331 |

## FAQ

### Is Deepseek Coder v2 better than Qwen3 32B?

Qwen3 32B is the stronger model overall, scoring 39.2 to 35.9 on the Noometry Index.

### Is Deepseek Coder v2 or Qwen3 32B better for coding?

They score almost the same on coding (38.1 vs 37.7); test both on your own repository before choosing.

### How many benchmarks do Deepseek Coder v2 and Qwen3 32B share?

13 benchmarks have published results for both models. Deepseek Coder v2 has 24 scored results on Noometry and Qwen3 32B has 26.