recover torchao

Author: Gasoonjia

.ci/docker/ci_commit_pins/pytorch.txt

@@ -1 +1 @@
-4d4abec80f03cd8fdefe1d9cb3a60d3690cd777e
+53a2908a10f414a2f85caa06703a26a40e873869

install_requirements.py

@@ -231,7 +231,6 @@ def install_requirements(use_pytorch_nightly):
         # that we don't need to set any version number there because they have already
         # been installed on CI before this step, so pip won't reinstall them
         f"torch==2.10.0.{NIGHTLY_VERSION}" if use_pytorch_nightly else "torch",
-        f"torchao==0.14.0{NIGHTLY_VERSION}" if use_pytorch_nightly else "torchao",
     ]
 
     # Install the requirements for core ExecuTorch package.
@@ -253,7 +252,7 @@ def install_requirements(use_pytorch_nightly):
     )
 
     LOCAL_REQUIREMENTS = [
-        # "third-party/ao",  # We need the latest kernels for fast iteration, so not relying on pypi.
+        "third-party/ao",  # We need the latest kernels for fast iteration, so not relying on pypi.
     ] + (
         [
             "extension/llm/tokenizers",  # TODO(larryliu0820): Setup a pypi package for this.