diff --git a/.github/workflows/build.yml b/.github/workflows/build.yml
index c595a3b6..1f6b2758 100644
--- a/.github/workflows/build.yml
+++ b/.github/workflows/build.yml
@@ -1,12 +1,11 @@
 name: Build and test cpu
 on:
   push:
-    branch: 'master'
     paths-ignore:
       - '**.md'
       - 'LICENSE'
   pull_request:
-    paths-ignore:
+    paths:
       - '**.md'
       - 'LICENSE'
 
diff --git a/.github/workflows/clang-format-check.yml b/.github/workflows/clang-format-check.yml
index dfd15b07..257da404 100644
--- a/.github/workflows/clang-format-check.yml
+++ b/.github/workflows/clang-format-check.yml
@@ -1,12 +1,11 @@
 name: clang-format Check
 on:
   push:
-    branch: 'master'
     paths-ignore:
       - '**.md'
       - 'LICENSE'
   pull_request:
-    paths-ignore:
+    paths:
       - '**.md'
       - 'LICENSE'
 
diff --git a/pyinfinitensor/pyproject.toml b/pyinfinitensor/pyproject.toml
index 8c438812..5b09e91e 100644
--- a/pyinfinitensor/pyproject.toml
+++ b/pyinfinitensor/pyproject.toml
@@ -12,7 +12,7 @@ requires-python = ">=3.7"
 keywords = ["optimizer"]
 license = { text = "Apache" }
 classifiers = ["Programming Language :: Python :: 3"]
-dependencies = ["onnx"]
+dependencies = ["onnx","onnx-simplifier"]
 
 [tool.setuptools.packages.find]
 where = ["src"]
diff --git a/pyinfinitensor/src/pyinfinitensor/onnx.py b/pyinfinitensor/src/pyinfinitensor/onnx.py
index af1e1f95..103af8e4 100644
--- a/pyinfinitensor/src/pyinfinitensor/onnx.py
+++ b/pyinfinitensor/src/pyinfinitensor/onnx.py
@@ -20,11 +20,14 @@ from onnx.checker import (
     check_node,
     check_value_info,
     check_tensor,
+    ValidationError,
 )
 from onnx.shape_inference import infer_shapes
 from onnx.numpy_helper import to_array
 from typing import Dict, List, Any, Tuple, Sequence, Union, Optional
 from functools import reduce
+from onnxsim import simplify
+import copy
 
 
 class OnnxStub:
@@ -33,6 +36,14 @@ class OnnxStub:
     It can be generated from an Onnx model object.
     """
     def __init__(self, model: ModelProto, runtime):
+        # We use some user-defined operators for distributed inference
+        try:
+            # onnx simplifier performs inplace simplify
+            model_simp, check = simplify(copy.deepcopy(model))
+            if check:
+                model = model_simp
+        except ValidationError:
+            pass
         self.inputs: Dict[str, backend.Tensor] = {}
         self.outputs: Dict[str, backend.Tensor] = {}
         self.initializer: Dict[int, TensorProto] = {}