llm-torch-profiler-analysis

1"""Backwards-compatibility shim for the unified LLM torch-profiler entrypoint.2 3The real implementation now lives in ``analyze_llm_torch_profile`` because this4skill covers SGLang, vLLM, TensorRT-LLM, and TokenSpeed. Older scripts and runbooks that5still invoke ``analyze_sglang_torch_profile.py`` keep working by forwarding to6that module.7"""8 9from __future__ import annotations10 11import sys12 13from analyze_llm_torch_profile import main14 15if __name__ == "__main__":16    raise SystemExit(main(sys.argv[1:]))17