Model comparison
Deepseek Coder v2 vs Qwen3.5 35B-A3B
Qwen3.5 35B-A3B is the stronger model overall, scoring 42.0 to 35.9 on the Noometry Index.
Last verified . 17 shared benchmarks.
DeepSeek
Alibaba (Qwen)Summary
- They share 17 benchmarks with published results for both. Deepseek Coder v2 scores higher in 1 category and Qwen3.5 35B-A3B in 7 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where Qwen3.5 35B-A3B leads 57.9 to 38.2.
Side by side
| Deepseek Coder v2 | Qwen3.5 35B-A3B | |
|---|---|---|
| Provider | DeepSeek | Alibaba (Qwen) |
| Noometry Index | 35.9 | 42.0 |
| Released | 2024-06-17 | 2026-02-01 |
| Weights | Open | Open |
| Context window | — | 262K |
| Max output | — | 66K |
| Input $ / M tokens | — | $0.25 |
| Output $ / M tokens | — | $2 |
| Results tracked | 24 | 28 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Deepseek Coder v2 leads
Deepseek Coder v2: 38.1 (#183), Qwen3.5 35B-A3B: 33.8 (#251)
| Benchmark | Deepseek Coder v2 | Qwen3.5 35B-A3B |
|---|---|---|
| LMArena Coding | 1251 | 1410 |
| LMArena WebDev | — | 1254 |
| SciCode | — | 29.3% |
| BigCodeBench Instruct | 48.2% | — |
| BigCodeBench Complete | 59.7% | — |
| HumanEval+ | 82.3% | — |
| MBPP+ | 75.1% | — |
Reasoning Too close to call
Deepseek Coder v2: 23.6 (#176), Qwen3.5 35B-A3B: 24.6 (#161)
| Benchmark | Deepseek Coder v2 | Qwen3.5 35B-A3B |
|---|---|---|
| LMArena Hard Prompts | 1207 | 1400 |
| CritPt | — | 0.6% |
| Chess Puzzles | — | 10% |
| DTBench | — | 80% |
| LMCA | — | 29.5% |
| Epoch Capabilities Index | — | 142.52 |
| WinoGrande | 83.7% | — |
Math Qwen3.5 35B-A3B leads
Deepseek Coder v2: 34.9 (#190), Qwen3.5 35B-A3B: 39.9 (#97)
| Benchmark | Deepseek Coder v2 | Qwen3.5 35B-A3B |
|---|---|---|
| LMArena Math | 1241 | 1404 |
| MathArena Final-Answer Competitions | — | 56% |
| OTIS Mock AIME 2024-2025 | — | 70% |
| GSM8K | 94.5% | — |
Knowledge Qwen3.5 35B-A3B leads
Deepseek Coder v2: 32.3 (#212), Qwen3.5 35B-A3B: 47.8 (#79)
| Benchmark | Deepseek Coder v2 | Qwen3.5 35B-A3B |
|---|---|---|
| LMArena Expert | 1181 | 1408 |
| GPQA Diamond | — | 83.5% |
| Vectara Hallucination Rate | — | 10.5% |
| ARC (AI2) Challenge | 64.3% | — |
Multilingual Qwen3.5 35B-A3B leads
Deepseek Coder v2: 36.3 (#240), Qwen3.5 35B-A3B: 50.0 (#127)
| Benchmark | Deepseek Coder v2 | Qwen3.5 35B-A3B |
|---|---|---|
| LMArena Non-English | 1182 | 1378 |
| LMArena Chinese | 1201 | 1457 |
| LMArena French | 1185 | 1412 |
| LMArena German | 1164 | 1367 |
| LMArena Japanese | 1126 | 1325 |
| LMArena Korean | 1104 | 1356 |
| LMArena Russian | 1188 | 1376 |
| LMArena Spanish | 1153 | 1392 |
Instruction Following Qwen3.5 35B-A3B leads
Deepseek Coder v2: 61.7 (#242), Qwen3.5 35B-A3B: 72.8 (#128)
| Benchmark | Deepseek Coder v2 | Qwen3.5 35B-A3B |
|---|---|---|
| LMArena Instruction Following | 1180 | 1379 |
Long Context Qwen3.5 35B-A3B leads
Deepseek Coder v2: 37.0 (#224), Qwen3.5 35B-A3B: 42.4 (#127)
| Benchmark | Deepseek Coder v2 | Qwen3.5 35B-A3B |
|---|---|---|
| LMArena Longer Query | 1219 | 1389 |
Writing & Preference Qwen3.5 35B-A3B leads
Deepseek Coder v2: 38.2 (#253), Qwen3.5 35B-A3B: 57.9 (#124)
| Benchmark | Deepseek Coder v2 | Qwen3.5 35B-A3B |
|---|---|---|
| LMArena Text | 1191 | 1395 |
| LMArena Creative Writing | 1120 | 1346 |
| LMArena Multi-Turn | 1177 | 1390 |
Frequently asked questions
Is Deepseek Coder v2 better than Qwen3.5 35B-A3B?
Qwen3.5 35B-A3B is the stronger model overall, scoring 42.0 to 35.9 on the Noometry Index.
Is Deepseek Coder v2 or Qwen3.5 35B-A3B better for coding?
Deepseek Coder v2 scores higher on coding benchmarks: 38.1 versus 33.8 in the Noometry coding category.
How many benchmarks do Deepseek Coder v2 and Qwen3.5 35B-A3B share?
17 benchmarks have published results for both models. Deepseek Coder v2 has 24 scored results on Noometry and Qwen3.5 35B-A3B has 28.