Nvidia shrinks LLM memory 20x without changing model weights

Nvidia's KV Cache Transform Coding (KVTC) compresses LLM key-value cache by 20x without model changes, cutting GPU memory costs and time-to-first-token by up to 8x for multi-turn AI applications.
TMCnet

Breaking the 100M Token Limit: EverMind's MSA Architecture Achieves Efficient End-to-End Long-Term Memory for LLMs

This approach can be viewed as a memory plug-in for large models, providing a fresh perspective and direction for solving the ...

New KV cache compaction technique cuts LLM memory 50x without accuracy loss

MIT researchers developed Attention Matching, a KV cache compaction technique that compresses LLM memory by 50x in seconds — ...
Semiconductor Engineering

Freeing Up Near-Memory Capacity For Cache Using Compression Techniques In A Flat Hybrid-Memory Architecture

A technical paper titled “HMComp: Extending Near-Memory Capacity using Compression in Hybrid Memory” was published by researchers at Chalmers University of Technology and ZeroPoint Technologies.
Electronic Design

Hardware Compression Works at the Memory Cache Level

How lossless data compression can reduce memory and power requirements. How ZeroPoint’s compression technology differs from the competition. One can never have enough memory, and one way to get more ...