Model comparison
DeepSeek-V2 (MoE-236B, May 2024) vs Qwen2.5 72B Instruct
Qwen2.5 72B Instruct has enough public results to be ranked (#267); DeepSeek-V2 (MoE-236B, May 2024) does not yet, so treat this comparison as directional.
Last verified . 10 shared benchmarks.
DeepSeek
Alibaba (Qwen)Summary
- They share 10 benchmarks with published results for both. DeepSeek-V2 (MoE-236B, May 2024) scores higher in 1 category and Qwen2.5 72B Instruct 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 33.2.
Side by side
| DeepSeek-V2 (MoE-236B, May 2024) | Qwen2.5 72B Instruct | |
|---|---|---|
| Provider | DeepSeek | Alibaba (Qwen) |
| Noometry Index | 40.3 | 31.9 |
| Released | 2024-05-07 | 2024-09 |
| Weights | Open | Open |
| Context window | — | 131K |
| Max output | — | 8K |
| Input $ / M tokens | — | $1.40 |
| Output $ / M tokens | — | $5.60 |
| Results tracked | 10 | 43 |
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), Qwen2.5 72B Instruct: 33.2 (#260)
| Benchmark | DeepSeek-V2 (MoE-236B, May 2024) | Qwen2.5 72B Instruct |
|---|---|---|
| BigCodeBench Instruct | 48.9% | 45.8% |
| BigCodeBench Complete | 59.4% | 55.9% |
| WeirdML | — | 16% |
| LMArena Coding | — | 1292 |
Agentic & Tool Use Not comparable
DeepSeek-V2 (MoE-236B, May 2024): —, Qwen2.5 72B Instruct: 22.1 (#133)
| Benchmark | DeepSeek-V2 (MoE-236B, May 2024) | Qwen2.5 72B Instruct |
|---|---|---|
| TheAgentCompany | — | 5.7% |
| BALROG | — | 16.2% |
| METR Time Horizons | — | 35.8% |
Reasoning Not comparable
DeepSeek-V2 (MoE-236B, May 2024): —, Qwen2.5 72B Instruct: 22.3 (#199)
| Benchmark | DeepSeek-V2 (MoE-236B, May 2024) | Qwen2.5 72B Instruct |
|---|---|---|
| BIG-Bench Hard | 78.8% | 79.8% |
| Epoch Capabilities Index | 124.77 | 129 |
| HellaSwag | 87.1% | 84.8% |
| PIQA | 83.9% | 82.6% |
| WinoGrande | 86.3% | 82.3% |
| LMArena Hard Prompts | — | 1271 |
| DTBench | — | 62.9% |
| LMCA | — | 13.4% |
| ForecastBench | — | 57.5 |
Math Not comparable
DeepSeek-V2 (MoE-236B, May 2024): —, Qwen2.5 72B Instruct: 19.3 (#287)
| Benchmark | DeepSeek-V2 (MoE-236B, May 2024) | Qwen2.5 72B Instruct |
|---|---|---|
| OTIS Mock AIME 2024-2025 | — | 8.1% |
| Omni-MATH | — | 33% |
| LMArena Math | — | 1283 |
| MATH Level 5 | — | 63.2% |
Knowledge Not comparable
DeepSeek-V2 (MoE-236B, May 2024): —, Qwen2.5 72B Instruct: 27.0 (#253)
| Benchmark | DeepSeek-V2 (MoE-236B, May 2024) | Qwen2.5 72B Instruct |
|---|---|---|
| ARC (AI2) Challenge | 92.2% | 94.5% |
| MMLU | 78.4% | 85.3% |
| TriviaQA | 80% | 71.9% |
| GPQA Diamond | — | 49.1% |
| MMLU-Pro | — | 63.1% |
| Confabulations | — | 19.1% |
| GPQA (HELM) | — | 42.6% |
| LMArena Expert | — | 1245 |
Multilingual Not comparable
DeepSeek-V2 (MoE-236B, May 2024): —, Qwen2.5 72B Instruct: 41.0 (#213)
| Benchmark | DeepSeek-V2 (MoE-236B, May 2024) | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Non-English | — | 1252 |
| LMArena Chinese | — | 1272 |
| LMArena French | — | 1280 |
| LMArena German | — | 1234 |
| LMArena Japanese | — | 1180 |
| LMArena Korean | — | 1188 |
| LMArena Russian | — | 1264 |
| LMArena Spanish | — | 1256 |
Instruction Following Not comparable
DeepSeek-V2 (MoE-236B, May 2024): —, Qwen2.5 72B Instruct: 65.5 (#221)
| Benchmark | DeepSeek-V2 (MoE-236B, May 2024) | Qwen2.5 72B Instruct |
|---|---|---|
| IFEval | — | 80.6% |
| LMArena Instruction Following | — | 1254 |
Long Context Not comparable
DeepSeek-V2 (MoE-236B, May 2024): —, Qwen2.5 72B Instruct: 38.9 (#188)
| Benchmark | DeepSeek-V2 (MoE-236B, May 2024) | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Longer Query | — | 1282 |
Writing & Preference Not comparable
DeepSeek-V2 (MoE-236B, May 2024): —, Qwen2.5 72B Instruct: 46.7 (#215)
| Benchmark | DeepSeek-V2 (MoE-236B, May 2024) | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Text | — | 1269 |
| LMArena Creative Writing | — | 1221 |
| WildBench | — | 80.2% |
| LMArena Multi-Turn | — | 1272 |
Frequently asked questions
Is DeepSeek-V2 (MoE-236B, May 2024) better than Qwen2.5 72B Instruct?
Qwen2.5 72B Instruct has enough public results to be ranked (#267); DeepSeek-V2 (MoE-236B, May 2024) does not yet, so treat this comparison as directional.
Is DeepSeek-V2 (MoE-236B, May 2024) or Qwen2.5 72B Instruct better for coding?
DeepSeek-V2 (MoE-236B, May 2024) scores higher on coding benchmarks: 40.4 versus 33.2 in the Noometry coding category.
How many benchmarks do DeepSeek-V2 (MoE-236B, May 2024) and Qwen2.5 72B Instruct share?
10 benchmarks have published results for both models. DeepSeek-V2 (MoE-236B, May 2024) has 10 scored results on Noometry and Qwen2.5 72B Instruct has 43.