ZML/LLMD: Open-source LLM server for high-performance inference

Original: RT @steeve: We're releasing ZML/LLMD, our homegrown LLM server built on top of our homegrown high performance heterogeneous inference stack…

Source: x.com ↗