# Hy3-preview NVFP4 — 2-node vLLM TP=2 serving (clone of dspark fabric, Hy3 args) WORKER_HOST=192.168.1.75 WORKER_DIR=/home/jon/hy3-serve MASTER_ADDR=10.10.10.1 MASTER_PORT=25000 NCCL_IB_HCA=rocep1s0f1,roceP2p1s0f1 NCCL_SOCKET_IFNAME=enp1s0f1np1 NCCL_IB_GID_INDEX=5 NCCL_CROSS_NIC=1 NCCL_NET=IB NCCL_IB_DISABLE=0 NCCL_CUMEM_ENABLE=0 NCCL_IGNORE_CPU_AFFINITY=1 NCCL_DEBUG=WARN NCCL_NVLS_ENABLE=0 HF_CACHE=/home/jon/.cache/huggingface WORKER_HF_CACHE=/home/jon/.cache/huggingface HY3_MODEL_DIR=/home/jon/models/Hy3-NVFP4-final SERVED_MODEL_NAME=hy3-nvfp4 HY3_VLLM_IMAGE=vllm/vllm-openai:hy3-arm64-cu130 VLLM_HOST=0.0.0.0 VLLM_HOST_IP=10.10.10.1 WORKER_VLLM_HOST_IP=10.10.10.2 MAX_MODEL_LEN=262144 MAX_NUM_SEQS=32 MAX_NUM_BATCHED_TOKENS=16384 GPU_MEMORY_UTILIZATION=0.90