DeepSeek V3.2 Speciale vs Kimi K2-Instruct-0905
DeepSeek vs Moonshot AI — benchmarks, pricing, and capabilities side by side.
- •DeepSeek V3.2 Speciale has the higher intelligence index (89.9 vs 66)
| DeepSeek V3.2 Speciale | Kimi K2-Instruct-0905 | |
|---|---|---|
| Intelligence index | 89.9 | 66 |
| Developer | DeepSeek | Moonshot AI |
| Type | LLM | LLM |
| Access | Open weights | Open weights |
| Context window | 163,840 tokens | — |
| Input price | $0.29 / 1M | — |
| Output price | $0.43 / 1M | — |
| Speed | — | — |
| Released | December 1, 2025 | September 5, 2025 |
| Parameters | — | 1000000000000 |
| Input modalities | Text | — |
| Output modalities | Text | — |
Shared benchmarks
DeepSeek V3.2 Speciale
Kimi K2-Instruct-0905
AIME 2025
96.7
49.5
GPQA Diamond
87.1
75.1
Humanity’s Last Exam
26.1
4.7
LiveCodeBench
89.6
53.7
MMLU-Pro
86.3
81.1