Model comparison
DeepSeek-V2 (MoE-236B, May 2024) vs Qwen1.5-32B
Qwen1.5-32B has enough public results to be ranked (#293); DeepSeek-V2 (MoE-236B, May 2024) does not yet, so treat this comparison as directional.
Last verified . 3 shared benchmarks.
DeepSeek
Alibaba (Qwen)Summary
- They share 3 benchmarks with published results for both. DeepSeek-V2 (MoE-236B, May 2024) scores higher in 1 category and Qwen1.5-32B in 0 categories; one gap is clear of the uncertainty.
- The widest gap is in coding, where DeepSeek-V2 (MoE-236B, May 2024) leads 40.4 to 31.7.
- The biggest single-benchmark swing is BigCodeBench Complete: 59.4% for DeepSeek-V2 (MoE-236B, May 2024) and 42% for Qwen1.5-32B.
Side by side
| DeepSeek-V2 (MoE-236B, May 2024) | Qwen1.5-32B | |
|---|---|---|
| Provider | DeepSeek | Alibaba (Qwen) |
| Noometry Index | 40.3 | 30.5 |
| Released | 2024-05-07 | 2024-02-04 |
| Weights | Open | Open |
| Context window | — | — |
| Max output | — | — |
| Input $ / M tokens | — | — |
| Output $ / M tokens | — | — |
| Results tracked | 10 | 21 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding DeepSeek-V2 (MoE-236B, May 2024) leads
DeepSeek-V2 (MoE-236B, May 2024): 40.4 (#139), Qwen1.5-32B: 31.7 (#282)
| Benchmark | DeepSeek-V2 (MoE-236B, May 2024) | Qwen1.5-32B |
|---|---|---|
| BigCodeBench Instruct | 48.9% | 32.3% |
| BigCodeBench Complete | 59.4% | 42% |
| LMArena Coding | — | 1155 |
Reasoning Not comparable
DeepSeek-V2 (MoE-236B, May 2024): —, Qwen1.5-32B: 21.8 (#212)
| Benchmark | DeepSeek-V2 (MoE-236B, May 2024) | Qwen1.5-32B |
|---|---|---|
| LMArena Hard Prompts | — | 1130 |
| BIG-Bench Hard | 78.8% | — |
| Epoch Capabilities Index | 124.77 | — |
| HellaSwag | 87.1% | — |
| PIQA | 83.9% | — |
| WinoGrande | 86.3% | — |
Math Not comparable
DeepSeek-V2 (MoE-236B, May 2024): —, Qwen1.5-32B: 33.0 (#207)
| Benchmark | DeepSeek-V2 (MoE-236B, May 2024) | Qwen1.5-32B |
|---|---|---|
| LMArena Math | — | 1155 |
Knowledge Not comparable
DeepSeek-V2 (MoE-236B, May 2024): —, Qwen1.5-32B: 13.5 (#296)
| Benchmark | DeepSeek-V2 (MoE-236B, May 2024) | Qwen1.5-32B |
|---|---|---|
| MMLU | 78.4% | 74.4% |
| GPQA Diamond | — | 30.7% |
| LMArena Expert | — | 1126 |
| ARC (AI2) Challenge | 92.2% | — |
| TriviaQA | 80% | — |
Multilingual Not comparable
DeepSeek-V2 (MoE-236B, May 2024): —, Qwen1.5-32B: 31.4 (#259)
| Benchmark | DeepSeek-V2 (MoE-236B, May 2024) | Qwen1.5-32B |
|---|---|---|
| LMArena Non-English | — | 1106 |
| LMArena Chinese | — | 1177 |
| LMArena French | — | 1101 |
| LMArena German | — | 1058 |
| LMArena Japanese | — | 1027 |
| LMArena Korean | — | 1008 |
| LMArena Russian | — | 1073 |
| LMArena Spanish | — | 1089 |
Instruction Following Not comparable
DeepSeek-V2 (MoE-236B, May 2024): —, Qwen1.5-32B: 57.7 (#265)
| Benchmark | DeepSeek-V2 (MoE-236B, May 2024) | Qwen1.5-32B |
|---|---|---|
| LMArena Instruction Following | — | 1116 |
Long Context Not comparable
DeepSeek-V2 (MoE-236B, May 2024): —, Qwen1.5-32B: 34.7 (#246)
| Benchmark | DeepSeek-V2 (MoE-236B, May 2024) | Qwen1.5-32B |
|---|---|---|
| LMArena Longer Query | — | 1146 |
Writing & Preference Not comparable
DeepSeek-V2 (MoE-236B, May 2024): —, Qwen1.5-32B: 34.2 (#271)
| Benchmark | DeepSeek-V2 (MoE-236B, May 2024) | Qwen1.5-32B |
|---|---|---|
| LMArena Text | — | 1137 |
| LMArena Creative Writing | — | 1083 |
| LMArena Multi-Turn | — | 1140 |
Frequently asked questions
Is DeepSeek-V2 (MoE-236B, May 2024) better than Qwen1.5-32B?
Qwen1.5-32B has enough public results to be ranked (#293); DeepSeek-V2 (MoE-236B, May 2024) does not yet, so treat this comparison as directional.
Is DeepSeek-V2 (MoE-236B, May 2024) or Qwen1.5-32B better for coding?
DeepSeek-V2 (MoE-236B, May 2024) scores higher on coding benchmarks: 40.4 versus 31.7 in the Noometry coding category.
How many benchmarks do DeepSeek-V2 (MoE-236B, May 2024) and Qwen1.5-32B share?
3 benchmarks have published results for both models. DeepSeek-V2 (MoE-236B, May 2024) has 10 scored results on Noometry and Qwen1.5-32B has 21.