Model comparison
DeepSeek-V3.2-Speciale vs StarCoder 2 15B
DeepSeek-V3.2-Speciale has enough public results to be ranked (#159); StarCoder 2 15B does not yet, so treat this comparison as directional.
Last verified . 0 shared benchmarks.
DeepSeek
NVIDIASummary
- The widest gap is in coding, where DeepSeek-V3.2-Speciale leads 39.9 to 36.0.
Side by side
| DeepSeek-V3.2-Speciale | StarCoder 2 15B | |
|---|---|---|
| Provider | DeepSeek | NVIDIA |
| Noometry Index | 39.1 | 37.5 |
| Released | 2025-12-01 | 2024-02-20 |
| Weights | Open | Open |
| Context window | 128K | — |
| Max output | 128K | — |
| Input $ / M tokens | $0.58 | — |
| Output $ / M tokens | $1.68 | — |
| Results tracked | 3 | 7 |
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), StarCoder 2 15B: 36.0
| Benchmark | DeepSeek-V3.2-Speciale | StarCoder 2 15B |
|---|---|---|
| WeirdML | 46.7% | — |
| BigCodeBench Complete | — | 38.4% |
| HumanEval+ | — | 37.8% |
Reasoning Not comparable
DeepSeek-V3.2-Speciale: 32.1 (#65), StarCoder 2 15B: —
| Benchmark | DeepSeek-V3.2-Speciale | StarCoder 2 15B |
|---|---|---|
| SimpleBench | 52.6% | — |
| Epoch Capabilities Index | — | 104.61 |
| WinoGrande | — | 64.3% |
Math Not comparable
DeepSeek-V3.2-Speciale: —, StarCoder 2 15B: —
| Benchmark | DeepSeek-V3.2-Speciale | StarCoder 2 15B |
|---|---|---|
| GSM8K | — | 57.7% |
Knowledge Not comparable
DeepSeek-V3.2-Speciale: —, StarCoder 2 15B: —
| Benchmark | DeepSeek-V3.2-Speciale | StarCoder 2 15B |
|---|---|---|
| ARC (AI2) Challenge | — | 47.2% |
| MMLU | — | 64.1% |
Writing & Preference Not comparable
DeepSeek-V3.2-Speciale: 46.1 (#217), StarCoder 2 15B: —
| Benchmark | DeepSeek-V3.2-Speciale | StarCoder 2 15B |
|---|---|---|
| EQ-Bench Creative Writing | 1276 | — |
Frequently asked questions
Is DeepSeek-V3.2-Speciale better than StarCoder 2 15B?
DeepSeek-V3.2-Speciale has enough public results to be ranked (#159); StarCoder 2 15B does not yet, so treat this comparison as directional.
Is DeepSeek-V3.2-Speciale or StarCoder 2 15B better for coding?
DeepSeek-V3.2-Speciale scores higher on coding benchmarks: 39.9 versus 36.0 in the Noometry coding category.
How many benchmarks do DeepSeek-V3.2-Speciale and StarCoder 2 15B share?
0 benchmarks have published results for both models. DeepSeek-V3.2-Speciale has 3 scored results on Noometry and StarCoder 2 15B has 7.