concurrency: 1
command:
    - python3
    - -m
    - vllm.entrypoints.openai.api_server
    - --model
    - /model
    - --served-model-name
    - llm
    - --max-model-len
    - '245000'
    - --gpu-memory-utilization
    - '0.9'
    - --trust-remote-code
    - -tp
    - '4'
    - --max-num-seqs
    - '1'
    - --disable-log-requests
    - --disable-frontend-multiprocessing
    - --max-num-batched-tokens
    - '16384'
    - --enable-chunked-prefill
    - --max-seq-len-to-capture
    - '245000'
    - --enable-auto-tool-choice
    - --tool-call-parser
    - qwen3_coder
    - --reasoning-parser
    - qwen3
    - --enable-prefix-caching
    # multi_system 兼容: 合并多/乱序 system 消息到首位, 消除 "System message must
    # be at the beginning" 4xx (数据集 34 条多 system 请求, 原模板 raise 致 15 条 4xx)。
    # 模型目录只读, patch_ops.sh 把改后模板部署到 /workspace/ (见 Dockerfile WORKDIR)。
    - --chat-template
    - /workspace/chat_template_multi_system.jinja
env:
    - name: VLLM_ENGINE_ITERATION_TIMEOUT_S
      value: 3600
    # corex runtime paths (must match patch_ops.sh)
    - name: PYTHONPATH
      value: /usr/local/corex/lib64/python3/dist-packages
    - name: LD_LIBRARY_PATH
      value: /usr/local/corex/lib64:/usr/local/iluvatar/lib64:/usr/local/openmpi/lib
    - name: PATH
      value: /usr/local/corex/bin:/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:/usr/local/openmpi/bin
    # five GatedDeltaNet/MoE/attention optimizations (default ON; =0 to disable)
    - name: MOE_NATIVE
      value: '1'
    - name: CHUNK_PARALLEL
      value: '1'
    - name: PREFIX_FLASH
      value: '1'
    - name: RMSNORM_NATIVE
      value: '1'
    - name: LOOP1_NATIVE
      value: '1'