1395516 1395512 nsys profile --output /root/arle-decode-hidden-scratch/docs/trace-artifacts/2026-05-15-dsv4-deepep/nsys-single-decode-token-route-grouped/trace --force-overwrite=true --trace cuda,nvtx,osrt --capture-range=cudaProfilerApi --capture-range-end=stop --export=sqlite --kill=none /root/arle/target/release/infer --model-path /root/DeepSeek-V4-Flash --port 18175 --num-slots 1 --max-seq-len 4096 --mem-fraction-static 0.10 --kv-cache-dtype fp8 --deepseek-distributed-layers 43
1395526       1 /opt/nvidia/nsight-systems/2023.2.3/target-linux-x64/nsys --start-agent --session-name=profile-infer-1395516 --shm-name=quadd-create-agent-76c45f9a-ec1f-4897-b329-f7b08f3c7f43
1395601 1395516 /opt/nvidia/nsight-systems/2023.2.3/target-linux-x64/nsys-launcher /tmp/b0d7-c294-6dc1-cec2 91
1395602 1395601 /root/arle/target/release/infer --model-path /root/DeepSeek-V4-Flash --port 18175 --num-slots 1 --max-seq-len 4096 --mem-fraction-static 0.10 --kv-cache-dtype fp8 --deepseek-distributed-layers 43
1395603 1395601 nsys-tee 3 4 5 6
