[Frontend] [Core] perf: Automatically detect vLLM-tensorized model, update tensorizer to version 2.9.0 (#4208)

2024-05-13 17:57:07 -04:00
parent 0fca3cdcf2
commit 8bc68e198c
10 changed files with 259 additions and 523 deletions
--- a/setup.py
+++ b/setup.py
@@ -426,7 +426,7 @@ setup(
    install_requires=get_requirements(),
    ext_modules=ext_modules,
    extras_require={
-        "tensorizer": ["tensorizer==2.9.0"],
+        "tensorizer": ["tensorizer>=2.9.0"],
    },
    cmdclass={"build_ext": cmake_build_ext} if not _is_neuron() else {},
    package_data=package_data,