Model comparison
Deepseek Coder v2 vs Qwen3-4B
Deepseek Coder v2 is the stronger model overall, scoring 35.9 to 31.9 on the Noometry Index.
Last verified . 0 shared benchmarks.
DeepSeek
Alibaba (Qwen)Summary
- The widest gap is in math, where Deepseek Coder v2 leads 34.9 to 29.7.
Side by side
| Deepseek Coder v2 | Qwen3-4B | |
|---|---|---|
| Provider | DeepSeek | Alibaba (Qwen) |
| Noometry Index | 35.9 | 31.9 |
| Released | 2024-06-17 | 2025-04-29 |
| Weights | Open | Open |
| Context window | — | — |
| Max output | — | — |
| Input $ / M tokens | — | — |
| Output $ / M tokens | — | — |
| Results tracked | 24 | 6 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Not comparable
Deepseek Coder v2: 38.1 (#183), Qwen3-4B: —
| Benchmark | Deepseek Coder v2 | Qwen3-4B |
|---|---|---|
| BigCodeBench Instruct | 48.2% | — |
| LMArena Coding | 1251 | — |
| BigCodeBench Complete | 59.7% | — |
| HumanEval+ | 82.3% | — |
| MBPP+ | 75.1% | — |
Agentic & Tool Use Not comparable
Deepseek Coder v2: —, Qwen3-4B: 27.6 (#100)
| Benchmark | Deepseek Coder v2 | Qwen3-4B |
|---|---|---|
| Berkeley Function Calling Leaderboard | — | 35.7% |
Reasoning Deepseek Coder v2 leads
Deepseek Coder v2: 23.6 (#176), Qwen3-4B: 19.2 (#268)
| Benchmark | Deepseek Coder v2 | Qwen3-4B |
|---|---|---|
| Chess Puzzles | — | 4% |
| LMArena Hard Prompts | 1207 | — |
| WinoGrande | 83.7% | — |
Math Deepseek Coder v2 leads
Deepseek Coder v2: 34.9 (#190), Qwen3-4B: 29.7 (#240)
| Benchmark | Deepseek Coder v2 | Qwen3-4B |
|---|---|---|
| MathArena Final-Answer Competitions | — | 38.5% |
| OTIS Mock AIME 2024-2025 | — | 52.2% |
| LMArena Math | 1241 | — |
| GSM8K | 94.5% | — |
Knowledge Too close to call
Deepseek Coder v2: 32.3 (#212), Qwen3-4B: 33.0 (#208)
| Benchmark | Deepseek Coder v2 | Qwen3-4B |
|---|---|---|
| GPQA Diamond | — | 52.3% |
| Vectara Hallucination Rate | — | 5.7% |
| LMArena Expert | 1181 | — |
| ARC (AI2) Challenge | 64.3% | — |
Multilingual Not comparable
Deepseek Coder v2: 36.3 (#240), Qwen3-4B: —
| Benchmark | Deepseek Coder v2 | Qwen3-4B |
|---|---|---|
| LMArena Non-English | 1182 | — |
| LMArena Chinese | 1201 | — |
| LMArena French | 1185 | — |
| LMArena German | 1164 | — |
| LMArena Japanese | 1126 | — |
| LMArena Korean | 1104 | — |
| LMArena Russian | 1188 | — |
| LMArena Spanish | 1153 | — |
Instruction Following Not comparable
Deepseek Coder v2: 61.7 (#242), Qwen3-4B: —
| Benchmark | Deepseek Coder v2 | Qwen3-4B |
|---|---|---|
| LMArena Instruction Following | 1180 | — |
Long Context Not comparable
Deepseek Coder v2: 37.0 (#224), Qwen3-4B: —
| Benchmark | Deepseek Coder v2 | Qwen3-4B |
|---|---|---|
| LMArena Longer Query | 1219 | — |
Writing & Preference Not comparable
Deepseek Coder v2: 38.2 (#253), Qwen3-4B: —
| Benchmark | Deepseek Coder v2 | Qwen3-4B |
|---|---|---|
| LMArena Text | 1191 | — |
| LMArena Creative Writing | 1120 | — |
| LMArena Multi-Turn | 1177 | — |
Frequently asked questions
Is Deepseek Coder v2 better than Qwen3-4B?
Deepseek Coder v2 is the stronger model overall, scoring 35.9 to 31.9 on the Noometry Index.
How many benchmarks do Deepseek Coder v2 and Qwen3-4B share?
0 benchmarks have published results for both models. Deepseek Coder v2 has 24 scored results on Noometry and Qwen3-4B has 6.