Get the latest tech news

DeepSeek-v3.2-Exp

Contribute to deepseek-ai/DeepSeek-V3.2-Exp development by creating an account on GitHub.

As an intermediate step toward our next-generation architecture, V3.2-Exp builds upon V3.1-Terminus by introducing DeepSeek Sparse Attention—a sparse attention mechanism designed to explore and validate optimizations for training and inference efficiency in long-context scenarios. DeepSeek Sparse Attention (DSA) achieves fine-grained sparse attention for the first time, delivering substantial improvements in long-context training and inference efficiency while maintaining virtually identical model output quality. To rigorously evaluate the impact of introducing sparse attention, we deliberately aligned the training configurations of DeepSeek-V3.2-Exp with V3.1-Terminus.

Get the Android app