Large-scale applications, such as generative AI, recommendation systems, big data, and HPC systems, require large-capacity ...

The butterfly bypass from the RotorQuant paper: TurboQuant applies a d×d Walsh-Hadamard Transform (butterfly network with log₂(d) stages across all 128 dimensions). PlanarQuant/IsoQuant apply ...

Researchers at North Carolina State University have developed a new AI-assisted tool that helps computer architects boost ...

When Google unveiled TurboQuant on March 24, headlines declared the algorithm could slash AI memory use sixfold with zero ...

Adding water to Cache Energy’s cement pellets causes a chemical reaction that releases heat. The reaction is reversible, allowing the system to store heat as well. CACHE ENERGY More than two millennia ...

Running a 70-billion-parameter large language model for 512 concurrent users can consume 512 GB of cache memory alone, nearly four times the memory needed for the model weights themselves. Google on ...

Google’s TurboQuant has the internet joking about Pied Piper from HBO's "Silicon Valley." The compression algorithm promises ...

Even if you don’t know much about the inner workings of generative AI models, you probably know they need a lot of memory. Hence, it is currently almost impossible to buy a measly stick of RAM without ...

AI has a growing memory problem. Google thinks it's found the answer, and it doesn't require more or better hardware. Originally detailed in an April 2025 paper, TurboQuant is an advanced compression ...

President Trump this week complained that America’s allies hadn’t stepped up to help break the Iranian blockade of the Strait of Hormuz. “Some are very enthusiastic about it, and some aren’t,” he said ...

The research introduces a novel memory architecture called MSA (Memory Sparse Attention). Through a combination of the Memory Sparse Attention mechanism, Document-wise RoPE for extreme context ...

Nvidia researchers have introduced a new technique that dramatically reduces how much memory large language models need to track conversation history — by as much as 20x — without modifying the model ...