2026-06-25 17:36:43 +08:00
|
|
|
concurrency: 1
|
2026-06-23 17:17:22 +08:00
|
|
|
command:
|
|
|
|
|
- python3
|
|
|
|
|
- -m
|
|
|
|
|
- vllm.entrypoints.openai.api_server
|
|
|
|
|
- --model
|
|
|
|
|
- /model
|
|
|
|
|
- --served-model-name
|
|
|
|
|
- llm
|
|
|
|
|
- --max-model-len
|
2026-07-15 00:49:24 +08:00
|
|
|
- '245000'
|
2026-06-23 17:17:22 +08:00
|
|
|
- --gpu-memory-utilization
|
2026-06-26 13:27:52 +08:00
|
|
|
- '0.9'
|
2026-06-23 17:17:22 +08:00
|
|
|
- --trust-remote-code
|
|
|
|
|
- -tp
|
|
|
|
|
- '4'
|
|
|
|
|
- --max-num-seqs
|
|
|
|
|
- '1'
|
|
|
|
|
- --disable-log-requests
|
|
|
|
|
- --disable-frontend-multiprocessing
|
|
|
|
|
- --max-num-batched-tokens
|
2026-07-15 00:49:24 +08:00
|
|
|
- '16384'
|
2026-06-23 17:17:22 +08:00
|
|
|
- --enable-chunked-prefill
|
|
|
|
|
- --max-seq-len-to-capture
|
2026-07-15 00:49:24 +08:00
|
|
|
- '245000'
|
2026-06-23 17:17:22 +08:00
|
|
|
- --enable-auto-tool-choice
|
|
|
|
|
- --tool-call-parser
|
|
|
|
|
- qwen3_coder
|
|
|
|
|
- --reasoning-parser
|
|
|
|
|
- qwen3
|
2026-06-26 13:27:52 +08:00
|
|
|
- --enable-prefix-caching
|
2026-07-15 00:49:24 +08:00
|
|
|
- --chat-template
|
|
|
|
|
- /workspace/chat_template_multi_system.jinja
|
2026-06-23 17:17:22 +08:00
|
|
|
env:
|
|
|
|
|
- name: VLLM_ENGINE_ITERATION_TIMEOUT_S
|
2026-06-25 17:36:43 +08:00
|
|
|
value: 3600
|
2026-07-15 00:49:24 +08:00
|
|
|
- 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
|
|
|
|
|
- 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'
|
2026-07-15 01:32:21 +08:00
|
|
|
|