Think your GPU doesn't have enough memory? Nvidia's AI texture compression trick looks like a killer solution in this demo ...

"It's literally going to feel like a sci-fi movie," Figure AI founder Brett Adcock says.

At TC Disrupt 2025, the AI conversation goes deeper than just the latest models and discusses the difference between ...

A research team led by Prof. Yang Yuchao from the School of Electronic and Computer Engineering at Peking University Shenzhen ...

Hardware system that eliminates comparators, accelerates nonlinear sorting, and delivers massive gains in speed and energy ...

Mem0's architecture is designed to LLM memory and enhance consistency for more reliable agent performance in long conversations.

KIOXIA AiSAQ software makes it possible to meet these demands efficiently, enabling large-scale generative AI without being constrained by limited memory resources.

The real breakthroughs will come from making AI not just larger, but smarter. Memory is the missing link—and solving it will reshape the future of artificial intelligence.

EnCharge AI, a startup that raised $144 million to date, announced the EnCharge EN100, an AI accelerator built on analog in-memory computing.

In an ongoing effort to improve the usability of AI vector database searches within retrieval-augmented generation (RAG) ...