Skip to content

Home Archive Tags Series Search About

Post / 2025-08-04

kkkzoz

Fast On-device LLM Inference with NPUs

LLM-Inference Edge-Devices NPU

Doc no.: KKZ-TN-083
Date: 2025-08-04
GMT day: 2025:216
Est. read: 1 min

Deja Vu Contextual Sparsity for Efficient LLMs at Inference Time

LLM Preliminaries

Comments

© kkkzoz

Built with Tola and tola-theme-nasapunk