Comment Comment by HarHarVeryFunny MoE is from Google (Noam Shazeer)MTP is from MetaAnother DeepSeek advance that the west are copying is DeepSeek Sparse Attention (DSA) parent HarHarVeryFunny 4 months ago
MTP is from Meta
Another DeepSeek advance that the west are copying is DeepSeek Sparse Attention (DSA)