fromTheregister2 months agoOpenAI gpt-oss LLMs use MXFP4: smaller, faster, cheaperMXFP4 is a 4-bit floating point data type defined by the Open Compute Project, allowing massive compute savings compared to traditional data types used by LLMs.Artificial intelligence
DevOpsfromTheregister5 months agoGoogle details 1 MW IT rack plans exploiting EV supply chainGoogle is upgrading datacenter racks to support 1 MW power loads, essential for scaling AI processing capabilities.