Model comparison
DeepSeek-V3.2-Speciale vs Qwen2-72B
DeepSeek-V3.2-Speciale is the stronger model overall, scoring 39.7 to 30.0 on the Noometry Index.
Last verified . 1 shared benchmarks.
DeepSeek
Alibaba (Qwen)Summary
- They share 1 benchmark with published results for both. DeepSeek-V3.2-Speciale scores higher in 3 categories and Qwen2-72B in 0 categories; 3 gaps are clear of the uncertainty.
- The widest gap is in coding, where DeepSeek-V3.2-Speciale leads 40.4 to 29.1.
- The biggest single-benchmark swing is WeirdML: 46.7% for DeepSeek-V3.2-Speciale and 11.3% for Qwen2-72B.
Side by side
| DeepSeek-V3.2-Speciale | Qwen2-72B | |
|---|---|---|
| Provider | DeepSeek | Alibaba (Qwen) |
| Noometry Index | 39.7 | 30.0 |
| Released | 2025-12-01 | 2024-06-07 |
| Weights | Open | Open |
| Context window | 128K | — |
| Max output | 128K | — |
| Input $ / M tokens | $0.58 | — |
| Output $ / M tokens | $1.68 | — |
| Results tracked | 3 | 26 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding DeepSeek-V3.2-Speciale leads
DeepSeek-V3.2-Speciale: 40.4 (#140), Qwen2-72B: 29.1 (#310)
| Benchmark | DeepSeek-V3.2-Speciale | Qwen2-72B |
|---|---|---|
| WeirdML | 46.7% | 11.3% |
| BigCodeBench Instruct | — | 38.5% |
| LMArena Coding | — | 1196 |
| BigCodeBench Complete | — | 54% |
Agentic & Tool Use Not comparable
DeepSeek-V3.2-Speciale: —, Qwen2-72B: 17.0 (#146)
| Benchmark | DeepSeek-V3.2-Speciale | Qwen2-72B |
|---|---|---|
| TheAgentCompany | — | 1.1% |
| METR Time Horizons | — | 29.9% |
Reasoning DeepSeek-V3.2-Speciale leads
DeepSeek-V3.2-Speciale: 32.9 (#73), Qwen2-72B: 23.2 (#181)
| Benchmark | DeepSeek-V3.2-Speciale | Qwen2-72B |
|---|---|---|
| SimpleBench | 52.6% | — |
| LMArena Hard Prompts | — | 1191 |
| Epoch Capabilities Index | — | 125.28 |
Math Not comparable
DeepSeek-V3.2-Speciale: —, Qwen2-72B: 30.2 (#236)
| Benchmark | DeepSeek-V3.2-Speciale | Qwen2-72B |
|---|---|---|
| LMArena Math | — | 1235 |
| MATH Level 5 | — | 39.1% |
Knowledge Not comparable
DeepSeek-V3.2-Speciale: —, Qwen2-72B: 21.2 (#275)
| Benchmark | DeepSeek-V3.2-Speciale | Qwen2-72B |
|---|---|---|
| GPQA Diamond | — | 40.8% |
| LMArena Expert | — | 1171 |
| MMLU | — | 82.4% |
Multilingual Not comparable
DeepSeek-V3.2-Speciale: —, Qwen2-72B: 35.9 (#244)
| Benchmark | DeepSeek-V3.2-Speciale | 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-V3.2-Speciale: —, Qwen2-72B: 61.7 (#241)
| Benchmark | DeepSeek-V3.2-Speciale | Qwen2-72B |
|---|---|---|
| LMArena Instruction Following | — | 1181 |
Long Context Not comparable
DeepSeek-V3.2-Speciale: —, Qwen2-72B: 36.1 (#235)
| Benchmark | DeepSeek-V3.2-Speciale | Qwen2-72B |
|---|---|---|
| LMArena Longer Query | — | 1192 |
Writing & Preference DeepSeek-V3.2-Speciale leads
DeepSeek-V3.2-Speciale: 46.0 (#222), Qwen2-72B: 40.8 (#241)
| Benchmark | DeepSeek-V3.2-Speciale | Qwen2-72B |
|---|---|---|
| LMArena Text | — | 1203 |
| LMArena Creative Writing | — | 1181 |
| EQ-Bench Creative Writing | 1276 | — |
| LMArena Multi-Turn | — | 1196 |
Frequently asked questions
Is DeepSeek-V3.2-Speciale better than Qwen2-72B?
DeepSeek-V3.2-Speciale is the stronger model overall, scoring 39.7 to 30.0 on the Noometry Index.
Is DeepSeek-V3.2-Speciale or Qwen2-72B better for coding?
DeepSeek-V3.2-Speciale scores higher on coding benchmarks: 40.4 versus 29.1 in the Noometry coding category.
How many benchmarks do DeepSeek-V3.2-Speciale and Qwen2-72B share?
1 benchmark has published results for both models. DeepSeek-V3.2-Speciale has 3 scored results on Noometry and Qwen2-72B has 26.