Model comparison
DeepSeek-V3.1 vs Qwen1.5-72B
DeepSeek-V3.1 is the stronger model overall, scoring 42.2 to 30.3 on the Noometry Index.
Last verified . 17 shared benchmarks.
DeepSeek
Alibaba (Qwen)Summary
- They share 17 benchmarks with published results for both. DeepSeek-V3.1 scores higher in 8 categories and Qwen1.5-72B in 0 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where DeepSeek-V3.1 leads 43.0 to 11.5.
Side by side
| DeepSeek-V3.1 | Qwen1.5-72B | |
|---|---|---|
| Provider | DeepSeek | Alibaba (Qwen) |
| Noometry Index | 42.2 | 30.3 |
| Released | 2025-08-21 | 2024-02-04 |
| Weights | Open | Open |
| Context window | 164K | — |
| Max output | 8K | — |
| Input $ / M tokens | $0.25 | — |
| Output $ / M tokens | $0.95 | — |
| Results tracked | 27 | 22 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding DeepSeek-V3.1 leads
DeepSeek-V3.1: 39.8 (#140), Qwen1.5-72B: 31.4 (#272)
| Benchmark | DeepSeek-V3.1 | Qwen1.5-72B |
|---|---|---|
| LMArena Coding | 1417 | 1166 |
| WeirdML | 38.4% | — |
| BigCodeBench Instruct | — | 33.2% |
| BigCodeBench Complete | — | 40.3% |
| HumanEval+ | — | 59.1% |
| MBPP+ | — | 61.6% |
Reasoning DeepSeek-V3.1 leads
DeepSeek-V3.1: 27.3 (#109), Qwen1.5-72B: 21.2 (#201)
| Benchmark | DeepSeek-V3.1 | Qwen1.5-72B |
|---|---|---|
| LMArena Hard Prompts | 1417 | 1148 |
| SimpleBench | 40% | — |
| Kagi LLM Benchmark | 53.2% | — |
| DTBench | 82.7% | — |
| LMCA | 24.3% | — |
| Epoch Capabilities Index | 139.92 | — |
| ForecastBench | 58 | — |
Math DeepSeek-V3.1 leads
DeepSeek-V3.1: 36.1 (#125), Qwen1.5-72B: 30.7 (#205)
| Benchmark | DeepSeek-V3.1 | Qwen1.5-72B |
|---|---|---|
| LMArena Math | 1420 | 1164 |
Knowledge DeepSeek-V3.1 leads
DeepSeek-V3.1: 43.0 (#86), Qwen1.5-72B: 11.5 (#295)
| Benchmark | DeepSeek-V3.1 | Qwen1.5-72B |
|---|---|---|
| LMArena Expert | 1403 | 1136 |
| GPQA Diamond | — | 28.8% |
| Vectara Hallucination Rate | 5.5% | — |
Multilingual DeepSeek-V3.1 leads
DeepSeek-V3.1: 51.8 (#100), Qwen1.5-72B: 33.4 (#248)
| Benchmark | DeepSeek-V3.1 | Qwen1.5-72B |
|---|---|---|
| LMArena Non-English | 1400 | 1135 |
| LMArena Chinese | 1470 | 1186 |
| LMArena French | 1447 | 1158 |
| LMArena German | 1412 | 1084 |
| LMArena Japanese | 1374 | 1061 |
| LMArena Korean | 1337 | 1049 |
| LMArena Russian | 1405 | 1104 |
| LMArena Spanish | 1429 | 1110 |
Instruction Following DeepSeek-V3.1 leads
DeepSeek-V3.1: 74.1 (#105), Qwen1.5-72B: 59.6 (#251)
| Benchmark | DeepSeek-V3.1 | Qwen1.5-72B |
|---|---|---|
| LMArena Instruction Following | 1400 | 1142 |
Long Context DeepSeek-V3.1 leads
DeepSeek-V3.1: 36.3 (#227), Qwen1.5-72B: 35.0 (#238)
| Benchmark | DeepSeek-V3.1 | Qwen1.5-72B |
|---|---|---|
| LMArena Longer Query | 1422 | 1157 |
| Fiction.LiveBench | 52.8% | — |
Writing & Preference DeepSeek-V3.1 leads
DeepSeek-V3.1: 60.5 (#93), Qwen1.5-72B: 37.4 (#253)
| Benchmark | DeepSeek-V3.1 | Qwen1.5-72B |
|---|---|---|
| LMArena Text | 1420 | 1166 |
| LMArena Creative Writing | 1401 | 1137 |
| LMArena Multi-Turn | 1408 | 1160 |
| EQ-Bench Creative Writing | 1436 | — |
Frequently asked questions
Is DeepSeek-V3.1 better than Qwen1.5-72B?
DeepSeek-V3.1 is the stronger model overall, scoring 42.2 to 30.3 on the Noometry Index.
Is DeepSeek-V3.1 or Qwen1.5-72B better for coding?
DeepSeek-V3.1 scores higher on coding benchmarks: 39.8 versus 31.4 in the Noometry coding category.
How many benchmarks do DeepSeek-V3.1 and Qwen1.5-72B share?
17 benchmarks have published results for both models. DeepSeek-V3.1 has 27 scored results on Noometry and Qwen1.5-72B has 22.