Microsoft's BitNet challenges industry norms with a minimalist approach using ternary weights that require just 400MB of memory while performing competitively against larger models on standard ...

Memory requirements are the most obvious advantage of reducing the complexity of a model's internal weights. The BitNet b1.58 ...

The BitNet b1.58 2B4T model was developed by Microsoft's General Artificial Intelligence group and contains two billion parameters – internal values that enable the model to ...

Microsoft’s model BitNet b1.58 2B4T is available on Hugging Face but doesn’t run on GPU and requires a proprietary framework.

Microsoft (MSFT) researchers claim they’ve developed the largest-scale 1-bit AI model, also known as a “bitnet,” to date. Called BitNet b1.58 ...

Microsoft researchers have developed — and released — a hyper-efficient AI model that can run on CPUs, including Apple's M2.

A group of computer scientists at Microsoft Research, working with a colleague from the University of Chinese Academy of ...

Bitnet works by simplifying the internal architecture of AI models. Instead of relying on full-precision or multi-bit quantization for their weights - the parameters that define the model's behavior - ...

Microsoft’s new BitNet b1.58 model significantly reduces memory and energy requirements while matching the capabilities of full-precision AI models, offering a promising low-resource alternative.

Yuliya Chernova reports on venture capital funds and startups for WSJ Pro Venture Capital out of New York. She’s interested ...