Model comparison
DeepSeek-V3.2-Speciale vs Qwen1.5-7B
DeepSeek-V3.2-Speciale is the stronger model overall, scoring 39.1 to 30.8 on the Noometry Index.
Last verified . 0 shared benchmarks.
DeepSeek
Alibaba (Qwen)Summary
- The widest gap is in writing & preference, where DeepSeek-V3.2-Speciale leads 46.1 to 29.8.
Side by side
| DeepSeek-V3.2-Speciale | Qwen1.5-7B | |
|---|---|---|
| Provider | DeepSeek | Alibaba (Qwen) |
| Noometry Index | 39.1 | 30.8 |
| Released | 2025-12-01 | 2024-02-04 |
| Weights | Open | Open |
| Context window | 128K | — |
| Max output | 128K | — |
| Input $ / M tokens | $0.58 | — |
| Output $ / M tokens | $1.68 | — |
| Results tracked | 3 | 13 |
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: 39.9 (#136), Qwen1.5-7B: 31.6 (#271)
| Benchmark | DeepSeek-V3.2-Speciale | Qwen1.5-7B |
|---|---|---|
| WeirdML | 46.7% | — |
| LMArena Coding | — | 1107 |
Reasoning DeepSeek-V3.2-Speciale leads
DeepSeek-V3.2-Speciale: 32.1 (#65), Qwen1.5-7B: 19.5 (#237)
| Benchmark | DeepSeek-V3.2-Speciale | Qwen1.5-7B |
|---|---|---|
| SimpleBench | 52.6% | — |
| LMArena Hard Prompts | — | 1066 |
Math Not comparable
DeepSeek-V3.2-Speciale: —, Qwen1.5-7B: 29.0 (#222)
| Benchmark | DeepSeek-V3.2-Speciale | Qwen1.5-7B |
|---|---|---|
| LMArena Math | — | 1080 |
Knowledge Not comparable
DeepSeek-V3.2-Speciale: —, Qwen1.5-7B: 28.2 (#239)
| Benchmark | DeepSeek-V3.2-Speciale | Qwen1.5-7B |
|---|---|---|
| LMArena Expert | — | 1055 |
| MMLU | — | 62.6% |
Multilingual Not comparable
DeepSeek-V3.2-Speciale: —, Qwen1.5-7B: 28.6 (#266)
| Benchmark | DeepSeek-V3.2-Speciale | Qwen1.5-7B |
|---|---|---|
| LMArena Non-English | — | 1058 |
| LMArena Chinese | — | 1140 |
| LMArena Russian | — | 1006 |
Instruction Following Not comparable
DeepSeek-V3.2-Speciale: —, Qwen1.5-7B: 54.4 (#276)
| Benchmark | DeepSeek-V3.2-Speciale | Qwen1.5-7B |
|---|---|---|
| LMArena Instruction Following | — | 1058 |
Long Context Not comparable
DeepSeek-V3.2-Speciale: —, Qwen1.5-7B: 33.1 (#261)
| Benchmark | DeepSeek-V3.2-Speciale | Qwen1.5-7B |
|---|---|---|
| LMArena Longer Query | — | 1090 |
Writing & Preference DeepSeek-V3.2-Speciale leads
DeepSeek-V3.2-Speciale: 46.1 (#217), Qwen1.5-7B: 29.8 (#288)
| Benchmark | DeepSeek-V3.2-Speciale | Qwen1.5-7B |
|---|---|---|
| LMArena Text | — | 1084 |
| LMArena Creative Writing | — | 1035 |
| EQ-Bench Creative Writing | 1276 | — |
| LMArena Multi-Turn | — | 1061 |
Frequently asked questions
Is DeepSeek-V3.2-Speciale better than Qwen1.5-7B?
DeepSeek-V3.2-Speciale is the stronger model overall, scoring 39.1 to 30.8 on the Noometry Index.
Is DeepSeek-V3.2-Speciale or Qwen1.5-7B better for coding?
DeepSeek-V3.2-Speciale scores higher on coding benchmarks: 39.9 versus 31.6 in the Noometry coding category.
How many benchmarks do DeepSeek-V3.2-Speciale and Qwen1.5-7B share?
0 benchmarks have published results for both models. DeepSeek-V3.2-Speciale has 3 scored results on Noometry and Qwen1.5-7B has 13.