French startup ZML releases free inference server to break Nvidia hardware lock-in

French AI startup ZML released LLMD on July 8, 2026, a free inference server that runs LLMs across Nvidia, AMD, Google TPU, Apple Metal, and Intel Arc chips. The product targets enterprises and cloud providers seeking to mix hardware vendors rather than depend on a single supplier. Turing Award winner Yann LeCun endorsed the company. LLMD competes with commercial inference engines vLLM and SGLang.