Model comparison
Deepseek Coder v2 vs Qwen2.5 32B Instruct
Deepseek Coder v2 is the stronger model overall, scoring 35.9 to 30.1 on the Noometry Index.
Last verified . 2 shared benchmarks.
DeepSeek
Alibaba (Qwen)Summary
- They share 2 benchmarks with published results for both. Deepseek Coder v2 scores higher in 3 categories and Qwen2.5 32B Instruct in 1 category; 3 gaps are clear of the uncertainty.
- The widest gap is in math, where Deepseek Coder v2 leads 34.9 to 16.2.
- The biggest single-benchmark swing is BigCodeBench Complete: 59.7% for Deepseek Coder v2 and 52.3% for Qwen2.5 32B Instruct.
Side by side
| Deepseek Coder v2 | Qwen2.5 32B Instruct | |
|---|---|---|
| Provider | DeepSeek | Alibaba (Qwen) |
| Noometry Index | 35.9 | 30.1 |
| Released | 2024-06-17 | 2024-09 |
| Weights | Open | Open |
| Context window | — | 131K |
| Max output | — | 8K |
| Input $ / M tokens | — | $0.70 |
| Output $ / M tokens | — | $2.80 |
| Results tracked | 24 | 7 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Too close to call
Deepseek Coder v2: 38.1 (#183), Qwen2.5 32B Instruct: 38.7 (#169)
| Benchmark | Deepseek Coder v2 | Qwen2.5 32B Instruct |
|---|---|---|
| BigCodeBench Instruct | 48.2% | 45% |
| BigCodeBench Complete | 59.7% | 52.3% |
| LMArena Coding | 1251 | — |
| HumanEval+ | 82.3% | — |
| MBPP+ | 75.1% | — |
Reasoning Deepseek Coder v2 leads
Deepseek Coder v2: 23.6 (#176), Qwen2.5 32B Instruct: 19.2 (#266)
| Benchmark | Deepseek Coder v2 | Qwen2.5 32B Instruct |
|---|---|---|
| Chess Puzzles | — | 0% |
| LMArena Hard Prompts | 1207 | — |
| Epoch Capabilities Index | — | 128.52 |
| WinoGrande | 83.7% | — |
Math Deepseek Coder v2 leads
Deepseek Coder v2: 34.9 (#190), Qwen2.5 32B Instruct: 16.2 (#296)
| Benchmark | Deepseek Coder v2 | Qwen2.5 32B Instruct |
|---|---|---|
| OTIS Mock AIME 2024-2025 | — | 7.4% |
| LMArena Math | 1241 | — |
| MATH Level 5 | — | 56.1% |
| GSM8K | 94.5% | — |
Knowledge Deepseek Coder v2 leads
Deepseek Coder v2: 32.3 (#212), Qwen2.5 32B Instruct: 24.9 (#266)
| Benchmark | Deepseek Coder v2 | Qwen2.5 32B Instruct |
|---|---|---|
| GPQA Diamond | — | 46.1% |
| LMArena Expert | 1181 | — |
| ARC (AI2) Challenge | 64.3% | — |
Multilingual Not comparable
Deepseek Coder v2: 36.3 (#240), Qwen2.5 32B Instruct: —
| Benchmark | Deepseek Coder v2 | Qwen2.5 32B Instruct |
|---|---|---|
| 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), Qwen2.5 32B Instruct: —
| Benchmark | Deepseek Coder v2 | Qwen2.5 32B Instruct |
|---|---|---|
| LMArena Instruction Following | 1180 | — |
Long Context Not comparable
Deepseek Coder v2: 37.0 (#224), Qwen2.5 32B Instruct: —
| Benchmark | Deepseek Coder v2 | Qwen2.5 32B Instruct |
|---|---|---|
| LMArena Longer Query | 1219 | — |
Writing & Preference Not comparable
Deepseek Coder v2: 38.2 (#253), Qwen2.5 32B Instruct: —
| Benchmark | Deepseek Coder v2 | Qwen2.5 32B Instruct |
|---|---|---|
| LMArena Text | 1191 | — |
| LMArena Creative Writing | 1120 | — |
| LMArena Multi-Turn | 1177 | — |
Frequently asked questions
Is Deepseek Coder v2 better than Qwen2.5 32B Instruct?
Deepseek Coder v2 is the stronger model overall, scoring 35.9 to 30.1 on the Noometry Index.
Is Deepseek Coder v2 or Qwen2.5 32B Instruct better for coding?
They score almost the same on coding (38.1 vs 38.7); test both on your own repository before choosing.
How many benchmarks do Deepseek Coder v2 and Qwen2.5 32B Instruct share?
2 benchmarks have published results for both models. Deepseek Coder v2 has 24 scored results on Noometry and Qwen2.5 32B Instruct has 7.