Inference Node — runs on each GPU machine, cuBLAS SGEMM via Panama FFI, pipeline forward pass, gRPC activation transfer

Tokenizer — DJL SentencePiece wrapper, chat template formatter, BPE support for LLaMA 2/3, Mistral, Gemma

Coordinator and Scheduler — continuous batching, leader election, gRPC streaming, Java 25 virtual threads

Juno-master standalone coordinator entry point. Full-cluster tests.