← Back to Model Beat
10Models·Apr 24

DeepSeek-V4: a million-token context that agents can actually use

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

Covered by 2 sources

Related stories

ModelsChina’s DeepSeek unveils long-awaited V4 AI model - Silicon RepublicApr 24 · 31 sourcesModelsIntroducing GPT-5.5Apr 23ModelsGPT-5.5 System CardApr 23ModelsScaling Codex to enterprises worldwideApr 21