DeepSeek V3.2 Speciale vs Qwen2.5 72B Instruct
DeepSeek vs Alibaba — benchmarks, pricing, and capabilities side by side.
- •DeepSeek V3.2 Speciale has the higher intelligence index (89.9 vs 59.1)
- •DeepSeek V3.2 Speciale is cheaper ($0.29 vs $0.36 per 1M input)
- •DeepSeek V3.2 Speciale has a larger context window (164K)
| DeepSeek V3.2 Speciale | Qwen2.5 72B Instruct | |
|---|---|---|
| Intelligence index | 89.9 | 59.1 |
| Developer | DeepSeek | Alibaba |
| Type | LLM | LLM |
| Access | Open weights | Open weights |
| Context window | 163,840 tokens | 131,072 tokens |
| Input price | $0.29 / 1M | $0.36 / 1M |
| Output price | $0.43 / 1M | $0.40 / 1M |
| Speed | — | 100 tok/s |
| Released | December 1, 2025 | September 19, 2024 |
| Parameters | — | — |
| Input modalities | Text | Text |
| Output modalities | Text | Text |
Shared benchmarks
DeepSeek V3.2 Speciale
Qwen2.5 72B Instruct
AIME 2025
96.7
14
GPQA Diamond
87.1
49
Humanity’s Last Exam
26.1
4.2
LiveCodeBench
89.6
55.5
MMLU-Pro
86.3
71.1