DeepSeek V3.2 Speciale vs DeepSeek-V4-Pro
DeepSeek vs DeepSeek — benchmarks, pricing, and capabilities side by side.
- •DeepSeek V3.2 Speciale has the higher intelligence index (89.9 vs 88.2)
- •DeepSeek V3.2 Speciale is cheaper ($0.29 vs $0.44 per 1M input)
- •DeepSeek-V4-Pro has a larger context window (1M)
| DeepSeek V3.2 Speciale | DeepSeek-V4-Pro | |
|---|---|---|
| Intelligence index | 89.9 | 88.2 |
| Developer | DeepSeek | DeepSeek |
| Type | LLM | LLM |
| Access | Open weights | Open weights |
| Context window | 163,840 tokens | 1,048,576 tokens |
| Input price | $0.29 / 1M | $0.44 / 1M |
| Output price | $0.43 / 1M | $0.87 / 1M |
| Speed | — | 30 tok/s |
| Released | December 1, 2025 | April 24, 2026 |
| Parameters | — | 1.6T (49B active) |
| Input modalities | Text | Text |
| Output modalities | Text | Text |
Shared benchmarks
DeepSeek V3.2 Speciale
DeepSeek-V4-Pro
GPQA Diamond
87.1
90.1
Humanity’s Last Exam
26.1
35.9
LiveCodeBench
89.6
93.5
MMLU-Pro
86.3
87.5