Skip to main content
Run any Skill in Manus
with one click
$pwd:

test-model-qwen3-8b-pd-nccl

// LightLLM Qwen3-8b PD disaggregation gsm8k: pd_master on 8089, prefill on 8001, decode on 8002, tp 2 each. Assign four GPUs by running nvidia-smi and deciding prefill/decode pairs (no fixed card IDs; no complex shell automation). lm_eval hits pd_master URL. HOST vs PD_MASTER_IP when co-located. Requires LOG_DIR, MODEL_DIR, proxy cleared, no_proxy, summary.txt. Use for PD NCCL-style separation tests.

$ git log --oneline --stat
stars:4,081
forks:332
updated:May 13, 2026 at 06:44
SKILL.md
readonly