Lightllm Profiler Control
DevOps & Securitylightllm-profiler-controlLightllm Profiler Control is an Agent Skill for devops & security published in the ModelTC/LightLLM repository on GitHub: a SKILL.md instruction file an AI agent can load to do this task.
LightLLM profiler 使用说明。用于需要启动或停止 LightLLM 的 torch_profiler / nvtx profiling 功能时,尤其是查看 --enable_profiling、/profiler_start、/profiler_stop 的使用方法。
QUICK START
How to use this skill
Bring this guide into your coding agent with a prompt tailored to the tool you use.
- Open your project in Codex.
- Copy and paste the prompt below.
- Review files and risks before approving any action.
Prompt to paste
Inspect the Agent Skill at https://github.com/ModelTC/LightLLM/blob/HEAD/skills/lightllm-profiler-control/SKILL.md and all supporting files. Report provenance, permissions, scripts, dependencies, and security risks. Propose an exact installation into .agents/skills/lightllm-profiler-control, but wait for my approval before writing files or running scripts.
Copying this prompt does not install or run the skill. Review third-party files before use.
LightLLM Profiler 使用说明
使用场景
当用户需要使用 LightLLM profiler 功能时使用本 skill,包括:
- 启动服务时打开 profiler 能力。
- 通过 HTTP 接口控制 profiler start / stop。
启动方式
服务启动时增加 --enable_profiling:
python -m lightllm.server.api_server \
--model_dir /path/to/model \
--enable_profiling torch_profiler
支持值:
torch_profiler:启用 PyTorch profiler,trace 默认写入./trace,也可通过LIGHTLLM_TRACE_DIR指定目录。nvtx:启用 NVTX range,配合 NVIDIA Nsight Systems 等外部工具采集。
未设置 --enable_profiling 时,/profiler_start 和 /profiler_stop 会返回未启用提示。
HTTP 控制接口
启动 profiler:
curl http://127.0.0.1:8000/profiler_start
停止 profiler:
curl http://127.0.0.1:8000/profiler_stop
端口 8000 替换为服务启动时的 --port。