DeepSeek V3.2 Exp Thinking
DeepSeek · text → text
The model DeepSeek-V3.2-Exp-Think is officially named deepseek-reasoner. It is an experimental version. As an intermediate step towards the next-generation architecture, V3.2-Exp introduces DeepSeek Sparse Attention (a sparse attention mechanism) based on V3.1-Terminus, exploring and validating exploratory optimizations for training and inference efficiency on long texts.
Input$0.274 /M
Output$0.411 /M
Cache read$0.0274 /M