Model comparison
DeepSeek Coder 33B vs Qwen2-72B
Qwen2-72B has enough public results to be ranked (#300); DeepSeek Coder 33B does not yet, so treat this comparison as directional.
Last verified . 4 shared benchmarks.
DeepSeek
Alibaba (Qwen)Summary
- They share 4 benchmarks with published results for both. DeepSeek Coder 33B scores higher in 1 category and Qwen2-72B in 0 categories; one gap is clear of the uncertainty.
- The widest gap is in coding, where DeepSeek Coder 33B leads 38.0 to 29.1.
Side by side
| DeepSeek Coder 33B | Qwen2-72B | |
|---|---|---|
| Provider | DeepSeek | Alibaba (Qwen) |
| Noometry Index | 38.9 | 30.0 |
| Released | 2023-11-02 | 2024-06-07 |
| Weights | Open | Open |
| Context window | — | — |
| Max output | — | — |
| Input $ / M tokens | — | — |
| Output $ / M tokens | — | — |
| Results tracked | 9 | 26 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding DeepSeek Coder 33B leads
DeepSeek Coder 33B: 38.0 (#184), Qwen2-72B: 29.1 (#310)
| Benchmark | DeepSeek Coder 33B | Qwen2-72B |
|---|---|---|
| BigCodeBench Instruct | 42% | 38.5% |
| BigCodeBench Complete | 51.1% | 54% |
| WeirdML | — | 11.3% |
| LMArena Coding | — | 1196 |
| HumanEval+ | 75% | — |
| MBPP+ | 70.1% | — |
Agentic & Tool Use Not comparable
DeepSeek Coder 33B: —, Qwen2-72B: 17.0 (#146)
| Benchmark | DeepSeek Coder 33B | Qwen2-72B |
|---|---|---|
| TheAgentCompany | — | 1.1% |
| METR Time Horizons | — | 29.9% |
Reasoning Not comparable
DeepSeek Coder 33B: —, Qwen2-72B: 23.2 (#181)
| Benchmark | DeepSeek Coder 33B | Qwen2-72B |
|---|---|---|
| Epoch Capabilities Index | 96.32 | 125.28 |
| LMArena Hard Prompts | — | 1191 |
| WinoGrande | 62% | — |
Math Not comparable
DeepSeek Coder 33B: —, Qwen2-72B: 30.2 (#236)
| Benchmark | DeepSeek Coder 33B | Qwen2-72B |
|---|---|---|
| LMArena Math | — | 1235 |
| MATH Level 5 | — | 39.1% |
| GSM8K | 35.4% | — |
Knowledge Not comparable
DeepSeek Coder 33B: —, Qwen2-72B: 21.2 (#275)
| Benchmark | DeepSeek Coder 33B | Qwen2-72B |
|---|---|---|
| MMLU | 39.4% | 82.4% |
| GPQA Diamond | — | 40.8% |
| LMArena Expert | — | 1171 |
| ARC (AI2) Challenge | 42.2% | — |
Multilingual Not comparable
DeepSeek Coder 33B: —, Qwen2-72B: 35.9 (#244)
| Benchmark | DeepSeek Coder 33B | Qwen2-72B |
|---|---|---|
| LMArena Non-English | — | 1176 |
| LMArena Chinese | — | 1240 |
| LMArena French | — | 1170 |
| LMArena German | — | 1151 |
| LMArena Japanese | — | 1111 |
| LMArena Korean | — | 1083 |
| LMArena Russian | — | 1169 |
| LMArena Spanish | — | 1169 |
Instruction Following Not comparable
DeepSeek Coder 33B: —, Qwen2-72B: 61.7 (#241)
| Benchmark | DeepSeek Coder 33B | Qwen2-72B |
|---|---|---|
| LMArena Instruction Following | — | 1181 |
Long Context Not comparable
DeepSeek Coder 33B: —, Qwen2-72B: 36.1 (#235)
| Benchmark | DeepSeek Coder 33B | Qwen2-72B |
|---|---|---|
| LMArena Longer Query | — | 1192 |
Writing & Preference Not comparable
DeepSeek Coder 33B: —, Qwen2-72B: 40.8 (#241)
| Benchmark | DeepSeek Coder 33B | Qwen2-72B |
|---|---|---|
| LMArena Text | — | 1203 |
| LMArena Creative Writing | — | 1181 |
| LMArena Multi-Turn | — | 1196 |
Frequently asked questions
Is DeepSeek Coder 33B better than Qwen2-72B?
Qwen2-72B has enough public results to be ranked (#300); DeepSeek Coder 33B does not yet, so treat this comparison as directional.
Is DeepSeek Coder 33B or Qwen2-72B better for coding?
DeepSeek Coder 33B scores higher on coding benchmarks: 38.0 versus 29.1 in the Noometry coding category.
How many benchmarks do DeepSeek Coder 33B and Qwen2-72B share?
4 benchmarks have published results for both models. DeepSeek Coder 33B has 9 scored results on Noometry and Qwen2-72B has 26.