cd/entity/Compressed Sparse Attention 2· home› entities› Compressed Sparse Attention 2

grep -l @compressed sparse attention 2 /news/*.json | wc -l → 1

Compressed Sparse Attention 2

mentions 1 type Person feed RSS

// recent coverage 1 mentions

00:00

2026-09-10

mindstudio.ai

artificial-intelligence

DeepSeek V4.1 Flash Specs: KV Cache Compression Explained

DeepSeek AI's model card for DeepSeek V4.1 Flash reports a global KV cache footprint of roughly 890 bytes per token, about a quarter of the 4x-larger footprint of predecessor DeepSeek V4 Flash and a 4…

// co-occurs with top 7 entities

DeepSeek 1 DeepSeek V4.1 Flash 1 DeepSeek V4 Flash 1 DeepSeek V1 1 Terminal-Bench 2.1 1 DeepSWE v1.1 1 CyberGym 1

// topics top 5 topics

artificial intelligence 1 large language models 1 ai research 1 ai infrastructure 1 ai products 1