Sparse Attention for Million-Token Context: Building DeepSeek-V4's Hybrid Attention
Sparse Attention for Million-Token Context: Building DeepSeek-V4's Hybrid Attention
intermediate~5 hours9 notebooks
Curator of this Module
RD
Raj Dandekar
Checking access…
Learning Path
Article
1
Introduction2
Notebook 23
Notebook 34
Notebook 45
Notebook 56
Notebook 67
Notebook 78
Notebook 89
ConclusionCase Study
Certificate