cd/entity/Multi-head Latent Attention· home› entities› Multi-head Latent Attention

grep -l @multi-head latent attention /news/*.json | wc -l → 1

Multi-head Latent Attention

mentions 1 type Person feed RSS

// recent coverage 1 mentions

00:01

2026-07-30

pub.towardsai.net

large-language-models

Inside Kimi K3 Technical Report

Moonshot AI's Kimi K3 technical report details a 2.8-trillion-parameter model with 104 billion activated parameters, a 1-million-token context window, and native vision, achieving 2.5x scaling efficie…

// co-occurs with top 7 entities

Moonshot AI 1 Kimi K3 1 Kimi K2 1 DeepSeek V3 1 DeepSeek-V2 1 Kimi Delta Attention 1 Attention Residual 1

// topics top 3 topics

large language models 1 ai research 1 ai infrastructure 1