Hacker News new | ask | show | jobs
Understanding Multi-Head Latent Attention (From DeepSeek) (shreyansh26.github.io)
2 points by shreyansh26 143 days ago
1 comments
shreyansh26 143 days ago
A short deep-dive on Multi-Head Latent Attention (MLA) (from DeepSeek): intuition + math, then a walk from MHA → GQA → MQA → MLA, with PyTorch code and the fusion/absorption optimizations for KV-cache efficiency.
link