Model comparison
DeepSeek-V2 (MoE-236B, May 2024) vs Qwen1.5 4b Chat
Qwen1.5 4b Chat has enough public results to be ranked (#322); DeepSeek-V2 (MoE-236B, May 2024) does not yet, so treat this comparison as directional.
Last verified . 0 shared benchmarks.
DeepSeek
Alibaba (Qwen)Summary
- The widest gap is in coding, where DeepSeek-V2 (MoE-236B, May 2024) leads 40.4 to 29.1.
Side by side
| DeepSeek-V2 (MoE-236B, May 2024) | Qwen1.5 4b Chat | |
|---|---|---|
| Provider | DeepSeek | Alibaba (Qwen) |
| Noometry Index | 40.3 | 28.8 |
| Released | 2024-05-07 | — |
| Weights | Open | Open |
| Context window | — | — |
| Max output | — | — |
| Input $ / M tokens | — | — |
| Output $ / M tokens | — | — |
| Results tracked | 10 | 13 |
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 4b Chat: 29.1 (#308)
| Benchmark | DeepSeek-V2 (MoE-236B, May 2024) | Qwen1.5 4b Chat |
|---|---|---|
| BigCodeBench Instruct | 48.9% | — |
| LMArena Coding | — | 999 |
| BigCodeBench Complete | 59.4% | — |
Reasoning Not comparable
DeepSeek-V2 (MoE-236B, May 2024): —, Qwen1.5 4b Chat: 18.5 (#279)
| Benchmark | DeepSeek-V2 (MoE-236B, May 2024) | Qwen1.5 4b Chat |
|---|---|---|
| LMArena Hard Prompts | — | 976 |
| 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 4b Chat: 30.4 (#234)
| Benchmark | DeepSeek-V2 (MoE-236B, May 2024) | Qwen1.5 4b Chat |
|---|---|---|
| LMArena Math | — | 1026 |
Knowledge Not comparable
DeepSeek-V2 (MoE-236B, May 2024): —, Qwen1.5 4b Chat: 26.7 (#255)
| Benchmark | DeepSeek-V2 (MoE-236B, May 2024) | Qwen1.5 4b Chat |
|---|---|---|
| LMArena Expert | — | 980 |
| ARC (AI2) Challenge | 92.2% | — |
| MMLU | 78.4% | — |
| TriviaQA | 80% | — |
Multilingual Not comparable
DeepSeek-V2 (MoE-236B, May 2024): —, Qwen1.5 4b Chat: 24.1 (#290)
| Benchmark | DeepSeek-V2 (MoE-236B, May 2024) | Qwen1.5 4b Chat |
|---|---|---|
| LMArena Non-English | — | 979 |
| LMArena Chinese | — | 1024 |
| LMArena German | — | 902 |
| LMArena Russian | — | 952 |
Instruction Following Not comparable
DeepSeek-V2 (MoE-236B, May 2024): —, Qwen1.5 4b Chat: 49.0 (#300)
| Benchmark | DeepSeek-V2 (MoE-236B, May 2024) | Qwen1.5 4b Chat |
|---|---|---|
| LMArena Instruction Following | — | 978 |
Long Context Not comparable
DeepSeek-V2 (MoE-236B, May 2024): —, Qwen1.5 4b Chat: 30.1 (#290)
| Benchmark | DeepSeek-V2 (MoE-236B, May 2024) | Qwen1.5 4b Chat |
|---|---|---|
| LMArena Longer Query | — | 988 |
Writing & Preference Not comparable
DeepSeek-V2 (MoE-236B, May 2024): —, Qwen1.5 4b Chat: 23.8 (#309)
| Benchmark | DeepSeek-V2 (MoE-236B, May 2024) | Qwen1.5 4b Chat |
|---|---|---|
| LMArena Text | — | 997 |
| LMArena Creative Writing | — | 969 |
| LMArena Multi-Turn | — | 977 |
Frequently asked questions
Is DeepSeek-V2 (MoE-236B, May 2024) better than Qwen1.5 4b Chat?
Qwen1.5 4b Chat has enough public results to be ranked (#322); 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 4b Chat better for coding?
DeepSeek-V2 (MoE-236B, May 2024) scores higher on coding benchmarks: 40.4 versus 29.1 in the Noometry coding category.
How many benchmarks do DeepSeek-V2 (MoE-236B, May 2024) and Qwen1.5 4b Chat share?
0 benchmarks have published results for both models. DeepSeek-V2 (MoE-236B, May 2024) has 10 scored results on Noometry and Qwen1.5 4b Chat has 13.