# Deepseek Coder v2 vs Qwen3-30B-A3B

> Qwen3-30B-A3B is the stronger model overall, scoring 38.9 to 35.9 on the Noometry Index.

- Canonical page: https://noometry.com/compare/deepseek-coder-v2-vs-qwen3-30b-a3b
- Last updated: 2026-10-11
- Shared benchmarks: 17

## Summary

- They share 17 benchmarks with published results for both. Deepseek Coder v2 scores higher in 3 categories and Qwen3-30B-A3B in 5 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where Qwen3-30B-A3B leads 55.6 to 38.2.

## Snapshot

| | Deepseek Coder v2 | Qwen3-30B-A3B |
|---|---|---|
| Provider | DeepSeek | Alibaba (Qwen) |
| Noometry Index | 35.9 | 38.9 |
| Rank | 220 | 179 |
| Context | — | 41K |
| Input $/M | — | $0.12 |
| Output $/M | — | $0.50 |
| Weights | Open | Open |

## Coding

- Deepseek Coder v2: 38.1 (#183)
- Qwen3-30B-A3B: 37.5 (#194)

| Benchmark | Deepseek Coder v2 | Qwen3-30B-A3B |
|---|---|---|
| LMArena Coding | 1251 | 1416 |
| SciCode | — | 33.3% |
| WeirdML | — | 29.8% |
| BigCodeBench Instruct | 48.2% | — |
| BigCodeBench Complete | 59.7% | — |
| HumanEval+ | 82.3% | — |
| MBPP+ | 75.1% | — |

## Agentic & Tool Use

- Deepseek Coder v2: —
- Qwen3-30B-A3B: 29.8 (#82)

| Benchmark | Deepseek Coder v2 | Qwen3-30B-A3B |
|---|---|---|
| Berkeley Function Calling Leaderboard | — | 41.4% |

## Reasoning

- Deepseek Coder v2: 23.6 (#176)
- Qwen3-30B-A3B: 22.2 (#204)

| Benchmark | Deepseek Coder v2 | Qwen3-30B-A3B |
|---|---|---|
| LMArena Hard Prompts | 1207 | 1398 |
| Kagi LLM Benchmark | — | 54.9% |
| CritPt | — | 0.3% |
| Chess Puzzles | — | 8% |
| DTBench | — | 69.3% |
| LMCA | — | 22.4% |
| Epoch Capabilities Index | — | 139.63 |
| WinoGrande | 83.7% | — |

## Math

- Deepseek Coder v2: 34.9 (#190)
- Qwen3-30B-A3B: 37.4 (#157)

| Benchmark | Deepseek Coder v2 | Qwen3-30B-A3B |
|---|---|---|
| LMArena Math | 1241 | 1394 |
| MathArena Final-Answer Competitions | — | 47.8% |
| OTIS Mock AIME 2024-2025 | — | 70.3% |
| GSM8K | 94.5% | — |

## Knowledge

- Deepseek Coder v2: 32.3 (#212)
- Qwen3-30B-A3B: 41.8 (#105)

| Benchmark | Deepseek Coder v2 | Qwen3-30B-A3B |
|---|---|---|
| LMArena Expert | 1181 | 1396 |
| GPQA Diamond | — | 70.1% |
| Confabulations | — | 12.3% |
| ARC (AI2) Challenge | 64.3% | — |

## Multilingual

- Deepseek Coder v2: 36.3 (#240)
- Qwen3-30B-A3B: 49.5 (#132)

| Benchmark | Deepseek Coder v2 | Qwen3-30B-A3B |
|---|---|---|
| LMArena Non-English | 1182 | 1372 |
| LMArena Chinese | 1201 | 1433 |
| LMArena French | 1185 | 1418 |
| LMArena German | 1164 | 1380 |
| LMArena Japanese | 1126 | 1337 |
| LMArena Korean | 1104 | 1331 |
| LMArena Russian | 1188 | 1370 |
| LMArena Spanish | 1153 | 1404 |

## Instruction Following

- Deepseek Coder v2: 61.7 (#242)
- Qwen3-30B-A3B: 72.0 (#142)

| Benchmark | Deepseek Coder v2 | Qwen3-30B-A3B |
|---|---|---|
| LMArena Instruction Following | 1180 | 1363 |

## Long Context

- Deepseek Coder v2: 37.0 (#224)
- Qwen3-30B-A3B: 31.0 (#283)

| Benchmark | Deepseek Coder v2 | Qwen3-30B-A3B |
|---|---|---|
| LMArena Longer Query | 1219 | 1379 |
| Fiction.LiveBench | — | 40.6% |

## Writing & Preference

- Deepseek Coder v2: 38.2 (#253)
- Qwen3-30B-A3B: 55.6 (#143)

| Benchmark | Deepseek Coder v2 | Qwen3-30B-A3B |
|---|---|---|
| LMArena Text | 1191 | 1384 |
| LMArena Creative Writing | 1120 | 1317 |
| LMArena Multi-Turn | 1177 | 1378 |
| Short-Story Creative Writing | — | 75.3% |

## FAQ

### Is Deepseek Coder v2 better than Qwen3-30B-A3B?

Qwen3-30B-A3B is the stronger model overall, scoring 38.9 to 35.9 on the Noometry Index.

### Is Deepseek Coder v2 or Qwen3-30B-A3B better for coding?

They score almost the same on coding (38.1 vs 37.5); test both on your own repository before choosing.

### How many benchmarks do Deepseek Coder v2 and Qwen3-30B-A3B share?

17 benchmarks have published results for both models. Deepseek Coder v2 has 24 scored results on Noometry and Qwen3-30B-A3B has 32.