DeepSeek V3.2 Speciale vs Kimi Linear 48B A3B Instruct
DeepSeek vs Moonshot AI — benchmarks, pricing, and capabilities side by side.
- •DeepSeek V3.2 Speciale has the higher intelligence index (89.9 vs 43.5)
- •Kimi Linear 48B A3B Instruct is cheaper ($0.00 vs $0.29 per 1M input)
| DeepSeek V3.2 Speciale | Kimi Linear 48B A3B Instruct | |
|---|---|---|
| Intelligence index | 89.9 | 43.5 |
| Developer | DeepSeek | Moonshot AI |
| Type | LLM | LLM |
| Access | Open weights | — |
| Context window | 163,840 tokens | — |
| Input price | $0.29 / 1M | $0.00 / 1M |
| Output price | $0.43 / 1M | $0.00 / 1M |
| Speed | — | — |
| Released | December 1, 2025 | October 30, 2025 |
| Parameters | — | — |
| Input modalities | Text | — |
| Output modalities | Text | — |
Shared benchmarks
DeepSeek V3.2 Speciale
Kimi Linear 48B A3B Instruct
AIME 2025
96.7
36.3
GPQA Diamond
87.1
41.2
Humanity’s Last Exam
26.1
2.7
LiveCodeBench
89.6
37.8
MMLU-Pro
86.3
58.5