Model comparison
Deepseek Coder v2 vs Qwen Max
Deepseek Coder v2 is the stronger model overall, scoring 35.9 to 34.7 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 3 categories and Qwen Max in 5 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where Deepseek Coder v2 leads 34.9 to 22.3.
- Deepseek Coder v2 has downloadable open weights; the other is API-only.
Side by side
| Deepseek Coder v2 | Qwen Max | |
|---|---|---|
| Provider | DeepSeek | Alibaba (Qwen) |
| Noometry Index | 35.9 | 34.7 |
| Released | 2024-06-17 | 2024-04-03 |
| Weights | Open | Proprietary |
| Context window | — | 33K |
| Max output | — | 8K |
| Input $ / M tokens | — | $1.60 |
| Output $ / M tokens | — | $6.40 |
| Results tracked | 24 | 23 |
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), Qwen Max: 30.7 (#292)
| Benchmark | Deepseek Coder v2 | Qwen Max |
|---|---|---|
| LMArena Coding | 1251 | 1288 |
| Aider Polyglot | — | 21.8% |
| BigCodeBench Instruct | 48.2% | — |
| BigCodeBench Complete | 59.7% | — |
| HumanEval+ | 82.3% | — |
| MBPP+ | 75.1% | — |
Reasoning Qwen Max leads
Deepseek Coder v2: 23.6 (#176), Qwen Max: 25.1 (#151)
| Benchmark | Deepseek Coder v2 | Qwen Max |
|---|---|---|
| LMArena Hard Prompts | 1207 | 1269 |
| WinoGrande | 83.7% | — |
Math Deepseek Coder v2 leads
Deepseek Coder v2: 34.9 (#190), Qwen Max: 22.3 (#276)
| Benchmark | Deepseek Coder v2 | Qwen Max |
|---|---|---|
| LMArena Math | 1241 | 1275 |
| OTIS Mock AIME 2024-2025 | — | 16.1% |
| MATH Level 5 | — | 67.2% |
| FrontierMath (Feb 2025 set) | — | 1% |
| GSM8K | 94.5% | — |
Knowledge Deepseek Coder v2 leads
Deepseek Coder v2: 32.3 (#212), Qwen Max: 30.3 (#228)
| Benchmark | Deepseek Coder v2 | Qwen Max |
|---|---|---|
| LMArena Expert | 1181 | 1248 |
| GPQA Diamond | — | 56.1% |
| ARC (AI2) Challenge | 64.3% | — |
Multilingual Qwen Max leads
Deepseek Coder v2: 36.3 (#240), Qwen Max: 41.8 (#202)
| Benchmark | Deepseek Coder v2 | Qwen Max |
|---|---|---|
| LMArena Non-English | 1182 | 1263 |
| LMArena Chinese | 1201 | 1254 |
| LMArena French | 1185 | 1330 |
| LMArena German | 1164 | 1254 |
| LMArena Japanese | 1126 | 1205 |
| LMArena Korean | 1104 | 1142 |
| LMArena Russian | 1188 | 1274 |
| LMArena Spanish | 1153 | 1290 |
Instruction Following Qwen Max leads
Deepseek Coder v2: 61.7 (#242), Qwen Max: 66.5 (#208)
| Benchmark | Deepseek Coder v2 | Qwen Max |
|---|---|---|
| LMArena Instruction Following | 1180 | 1262 |
Long Context Qwen Max leads
Deepseek Coder v2: 37.0 (#224), Qwen Max: 39.4 (#180)
| Benchmark | Deepseek Coder v2 | Qwen Max |
|---|---|---|
| LMArena Longer Query | 1219 | 1288 |
| Fiction.LiveBench | — | 66.7% |
Writing & Preference Qwen Max leads
Deepseek Coder v2: 38.2 (#253), Qwen Max: 47.8 (#205)
| Benchmark | Deepseek Coder v2 | Qwen Max |
|---|---|---|
| LMArena Text | 1191 | 1282 |
| LMArena Creative Writing | 1120 | 1248 |
| LMArena Multi-Turn | 1177 | 1277 |
Frequently asked questions
Is Deepseek Coder v2 better than Qwen Max?
Deepseek Coder v2 is the stronger model overall, scoring 35.9 to 34.7 on the Noometry Index.
Is Deepseek Coder v2 or Qwen Max better for coding?
Deepseek Coder v2 scores higher on coding benchmarks: 38.1 versus 30.7 in the Noometry coding category.
How many benchmarks do Deepseek Coder v2 and Qwen Max share?
17 benchmarks have published results for both models. Deepseek Coder v2 has 24 scored results on Noometry and Qwen Max has 23.