diff --git a/pyproject.toml b/pyproject.toml
index bd028a41f04..be8776e3181 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -135,7 +135,6 @@ git_describe_command = [
     "--long",
     "--first-parent",
 ]
-version_scheme = "post-release"
 fallback_version = "0.0.0"
 [tool.hatch.metadata]
 allow-direct-references = true
diff --git a/src/bentoml/_internal/bento/build_config.py b/src/bentoml/_internal/bento/build_config.py
index 9f6fa5864eb..791e93331b2 100644
--- a/src/bentoml/_internal/bento/build_config.py
+++ b/src/bentoml/_internal/bento/build_config.py
@@ -580,7 +580,7 @@ def write_to_bento(
                 args.extend(pip_args)
             f.write(
                 self._jinja_environment.get_template("install.sh.j2").render(
-                    bentoml_version=clean_bentoml_version(BENTOML_VERSION),
+                    bentoml_version=clean_bentoml_version(),
                     pip_args=shlex.join(args),
                 )
             )
diff --git a/src/bentoml/_internal/configuration/__init__.py b/src/bentoml/_internal/configuration/__init__.py
index 05d21c6719e..24083233c8f 100644
--- a/src/bentoml/_internal/configuration/__init__.py
+++ b/src/bentoml/_internal/configuration/__init__.py
@@ -56,10 +56,10 @@ def expand_env_var(env_var: str) -> str:
             env_var = interpolated
 
 
-def clean_bentoml_version(bentoml_version: str) -> str:
-    post_version = ".".join(bentoml_version.split(".")[:3])
+@lru_cache(maxsize=1)
+def clean_bentoml_version() -> str:
     try:
-        version = Version(post_version)
+        version = Version(BENTOML_VERSION).base_version
         return str(version)
     except ValueError:
         raise BentoMLException("Errors while parsing BentoML version.") from None
diff --git a/src/bentoml/serving.py b/src/bentoml/serving.py
index 260e6016391..891bf888c33 100644
--- a/src/bentoml/serving.py
+++ b/src/bentoml/serving.py
@@ -1,6 +1,5 @@
 from __future__ import annotations
 
-import asyncio
 import contextlib
 import json
 import logging
@@ -683,8 +682,7 @@ def serve_grpc_production(
 
     close_child_stdin: bool = False if development_mode else True
 
-    loop = asyncio.get_event_loop()
-    loop.run_until_complete(on_service_deployment(svc))
+    on_service_deployment(svc)
 
     with contextlib.ExitStack() as port_stack:
         api_port = port_stack.enter_context(
diff --git a/src/bentoml/testing/pytest/plugin.py b/src/bentoml/testing/pytest/plugin.py
index 232a14e0f26..18604034bbb 100644
--- a/src/bentoml/testing/pytest/plugin.py
+++ b/src/bentoml/testing/pytest/plugin.py
@@ -12,7 +12,6 @@
 from pytest import MonkeyPatch
 
 import bentoml
-from bentoml._internal.configuration import BENTOML_VERSION
 from bentoml._internal.configuration import clean_bentoml_version
 from bentoml._internal.configuration.containers import BentoMLContainer
 from bentoml._internal.models import ModelContext
@@ -46,7 +45,7 @@
 
 @pytest.mark.tryfirst
 def pytest_report_header(config: Config) -> list[str]:
-    return [f"bentoml: version={clean_bentoml_version(BENTOML_VERSION)}"]
+    return [f"bentoml: version={clean_bentoml_version()}"]
 
 
 @pytest.hookimpl