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.

Open source Mamba 3 arrives to surpass Transformer architecture with nearly 4% improved language modeling, reduced latency

This release is good for developers building long-context applications, real-time reasoning agents, or those seeking to ...

Why some moments endure: Episodic memory encoding fluctuates with brain's theta rhythms

For almost a century, psychologists and neuroscientists have been trying to understand how humans memorize different types of ...

Nanoengineered spintronic device can store data in four different ways

Over the past decades, electronics engineers have been trying to develop increasingly smaller devices that can store ...

NVIDIA Corporation (NVDA) Presents at NVIDIA GTC AI Conference 2026 Prepared Remarks Transcript

Welcome to the stage, NVIDIA Founder and CEO, Jensen Huang. Welcome to GTC. I just want to remind you, this is a tech conference. All these people are lining up so early in the morning, all of you in ...