Model comparison
Deepseek Coder v2 vs DeepSeek-V3.2-Speciale
DeepSeek-V3.2-Speciale is the stronger model overall, scoring 39.7 to 35.9 on the Noometry Index.
Last verified . 0 shared benchmarks.
DeepSeekSummary
- The widest gap is in reasoning, where DeepSeek-V3.2-Speciale leads 32.9 to 23.6.
Side by side
| Deepseek Coder v2 | DeepSeek-V3.2-Speciale | |
|---|---|---|
| Provider | DeepSeek | DeepSeek |
| Noometry Index | 35.9 | 39.7 |
| Released | 2024-06-17 | 2025-12-01 |
| Weights | Open | Open |
| Context window | — | 128K |
| Max output | — | 128K |
| Input $ / M tokens | — | $0.58 |
| Output $ / M tokens | — | $1.68 |
| Results tracked | 24 | 3 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding DeepSeek-V3.2-Speciale leads
Deepseek Coder v2: 38.1 (#183), DeepSeek-V3.2-Speciale: 40.4 (#140)
| Benchmark | Deepseek Coder v2 | DeepSeek-V3.2-Speciale |
|---|---|---|
| WeirdML | — | 46.7% |
| BigCodeBench Instruct | 48.2% | — |
| LMArena Coding | 1251 | — |
| BigCodeBench Complete | 59.7% | — |
| HumanEval+ | 82.3% | — |
| MBPP+ | 75.1% | — |
Reasoning DeepSeek-V3.2-Speciale leads
Deepseek Coder v2: 23.6 (#176), DeepSeek-V3.2-Speciale: 32.9 (#73)
| Benchmark | Deepseek Coder v2 | DeepSeek-V3.2-Speciale |
|---|---|---|
| SimpleBench | — | 52.6% |
| LMArena Hard Prompts | 1207 | — |
| WinoGrande | 83.7% | — |
Math Not comparable
Deepseek Coder v2: 34.9 (#190), DeepSeek-V3.2-Speciale: —
| Benchmark | Deepseek Coder v2 | DeepSeek-V3.2-Speciale |
|---|---|---|
| LMArena Math | 1241 | — |
| GSM8K | 94.5% | — |
Knowledge Not comparable
Deepseek Coder v2: 32.3 (#212), DeepSeek-V3.2-Speciale: —
| Benchmark | Deepseek Coder v2 | DeepSeek-V3.2-Speciale |
|---|---|---|
| LMArena Expert | 1181 | — |
| ARC (AI2) Challenge | 64.3% | — |
Multilingual Not comparable
Deepseek Coder v2: 36.3 (#240), DeepSeek-V3.2-Speciale: —
| Benchmark | Deepseek Coder v2 | DeepSeek-V3.2-Speciale |
|---|---|---|
| LMArena Non-English | 1182 | — |
| LMArena Chinese | 1201 | — |
| LMArena French | 1185 | — |
| LMArena German | 1164 | — |
| LMArena Japanese | 1126 | — |
| LMArena Korean | 1104 | — |
| LMArena Russian | 1188 | — |
| LMArena Spanish | 1153 | — |
Instruction Following Not comparable
Deepseek Coder v2: 61.7 (#242), DeepSeek-V3.2-Speciale: —
| Benchmark | Deepseek Coder v2 | DeepSeek-V3.2-Speciale |
|---|---|---|
| LMArena Instruction Following | 1180 | — |
Long Context Not comparable
Deepseek Coder v2: 37.0 (#224), DeepSeek-V3.2-Speciale: —
| Benchmark | Deepseek Coder v2 | DeepSeek-V3.2-Speciale |
|---|---|---|
| LMArena Longer Query | 1219 | — |
Writing & Preference DeepSeek-V3.2-Speciale leads
Deepseek Coder v2: 38.2 (#253), DeepSeek-V3.2-Speciale: 46.0 (#222)
| Benchmark | Deepseek Coder v2 | DeepSeek-V3.2-Speciale |
|---|---|---|
| LMArena Text | 1191 | — |
| LMArena Creative Writing | 1120 | — |
| EQ-Bench Creative Writing | — | 1276 |
| LMArena Multi-Turn | 1177 | — |
Frequently asked questions
Is Deepseek Coder v2 better than DeepSeek-V3.2-Speciale?
DeepSeek-V3.2-Speciale is the stronger model overall, scoring 39.7 to 35.9 on the Noometry Index.
Is Deepseek Coder v2 or DeepSeek-V3.2-Speciale better for coding?
DeepSeek-V3.2-Speciale scores higher on coding benchmarks: 40.4 versus 38.1 in the Noometry coding category.
How many benchmarks do Deepseek Coder v2 and DeepSeek-V3.2-Speciale share?
0 benchmarks have published results for both models. Deepseek Coder v2 has 24 scored results on Noometry and DeepSeek-V3.2-Speciale has 3.