DeepSeek V3.2 Speciale vs MiniMax-M2
DeepSeek vs MiniMax — benchmarks, pricing, and capabilities side by side.
- •DeepSeek V3.2 Speciale has the higher intelligence index (89.9 vs 76)
- •MiniMax-M2 is cheaper ($0.26 vs $0.29 per 1M input)
- •MiniMax-M2 has a larger context window (205K)
| DeepSeek V3.2 Speciale | MiniMax-M2 | |
|---|---|---|
| Intelligence index | 89.9 | 76 |
| Developer | DeepSeek | MiniMax |
| Type | LLM | LLM |
| Access | Open weights | Open weights |
| Context window | 163,840 tokens | 204,800 tokens |
| Input price | $0.29 / 1M | $0.26 / 1M |
| Output price | $0.43 / 1M | $1.00 / 1M |
| Speed | — | 91 tok/s |
| Released | December 1, 2025 | October 27, 2025 |
| Parameters | — | 230B (10B active) |
| Input modalities | Text | Text |
| Output modalities | Text | Text |
Shared benchmarks
DeepSeek V3.2 Speciale
MiniMax-M2
AIME 2025
96.7
78.3
GPQA Diamond
87.1
77.7
Humanity’s Last Exam
26.1
12.5
LiveCodeBench
89.6
82.6
MMLU-Pro
86.3
82