ZML, a French AI startup, launched its LLM inference server, LLMD, designed to run open-source large language models on various chips including those from Nvidia, AMD, Google, Apple, and Intel, aiming for maximum speed and efficiency. The company, backed by $20 million in funding from investors like 20VC and Kima Ventures, seeks to break vendor lock-in and reduce AI-related costs by enabling enterprises to use a mix of hardware.