← Back to Model Beat
Models·Apr 24·all news from April 24, 2026

DeepSeek AI Releases DeepSeek-V4: Compressed Sparse Attention and Heavily Compressed Attention Enable One-Million-Token Contexts - MarkTechPost

DeepSeek AI Releases DeepSeek-V4: Compressed Sparse Attention and Heavily Compressed Attention Enable One-Million-Token Contexts MarkTechPost

Covered by 3 sources

Related stories

ModelsChina’s DeepSeek unveils long-awaited V4 AI model - Silicon RepublicApr 23 · 38 sourcesModelsIntroducing GPT-5.5Apr 23ModelsGPT-5.5 System CardApr 23ModelsDeepSeek Unveils Flagship AI Model a Year After Breakthrough - Bloomberg.comApr 24 · 11 sources