Multiverse Computing Unveils Breakthrough: All CompactifAI Models Now Run on Intel Xeon 6 Processors

Multiverse Computing announced its CompactifAI-compressed version of Llama 3.3 70B now runs on Intel Xeon 6 processors, delivering 93.6% improvement in output throughput and 48.6% latency reduction while maintaining 97% baseline accuracy. The compressed model reduces disk size by 50% and integrates with mainstream open-source frameworks, targeting enterprise applications in finance, healthcare, and manufacturing.