diff --git a/bentoml/_internal/configuration/v1/__init__.py b/bentoml/_internal/configuration/v1/__init__.py
index 5b76663e2b4..f06f630395f 100644
--- a/bentoml/_internal/configuration/v1/__init__.py
+++ b/bentoml/_internal/configuration/v1/__init__.py
@@ -4,6 +4,7 @@
 
 import schema as s
 
+from ....exceptions import BentoMLConfigException
 from ..helpers import ensure_range
 from ..helpers import flatten_dict
 from ..helpers import rename_fields
@@ -134,6 +135,13 @@
 
 
 def migrate_to_v2(*, override_config: dict[str, t.Any]):
+    try:
+        SCHEMA.validate(override_config)
+    except s.SchemaError as e:
+        raise BentoMLConfigException(
+            f"Invalid configuration file was given:\n{e}"
+        ) from None
+
     # We will use a flattened config to make it easier to migrate,
     # Then we will convert it back to a nested config.
     flatten = dict(flatten_dict(override_config))
diff --git a/bentoml/start.py b/bentoml/start.py
index 03d083b7c86..b02ee550528 100644
--- a/bentoml/start.py
+++ b/bentoml/start.py
@@ -3,7 +3,6 @@
 import os
 import sys
 import json
-import math
 import typing as t
 import logging
 import contextlib
@@ -122,7 +121,7 @@ def start_http_server(
     port: int = Provide[BentoMLContainer.api_server_config.port],
     host: str = Provide[BentoMLContainer.api_server_config.host],
     backlog: int = Provide[BentoMLContainer.api_server_config.backlog],
-    api_workers: int | None = None,
+    api_workers: int = Provide[BentoMLContainer.api_server_workers],
     ssl_certfile: str | None = Provide[BentoMLContainer.api_server_config.ssl.certfile],
     ssl_keyfile: str | None = Provide[BentoMLContainer.api_server_config.ssl.keyfile],
     ssl_keyfile_password: str
@@ -140,7 +139,6 @@ def start_http_server(
     from .serve import construct_ssl_args
     from .serve import PROMETHEUS_MESSAGE
     from .serve import ensure_prometheus_dir
-    from ._internal.resource import CpuResource
     from ._internal.utils.circus import create_standalone_arbiter
     from ._internal.utils.analytics import track_serve
 
@@ -200,7 +198,7 @@ def start_http_server(
                 ),
             ],
             working_dir=working_dir,
-            numprocesses=api_workers or math.ceil(CpuResource.from_system()),
+            numprocesses=api_workers,
         )
     )
     if BentoMLContainer.api_server_config.metrics.enabled.get():
@@ -235,7 +233,7 @@ def start_grpc_server(
     port: int = Provide[BentoMLContainer.grpc.port],
     host: str = Provide[BentoMLContainer.grpc.host],
     backlog: int = Provide[BentoMLContainer.api_server_config.backlog],
-    api_workers: int | None = None,
+    api_workers: int = Provide[BentoMLContainer.api_server_workers],
     reflection: bool = Provide[BentoMLContainer.grpc.reflection.enabled],
     max_concurrent_streams: int
     | None = Provide[BentoMLContainer.grpc.max_concurrent_streams],
@@ -247,7 +245,6 @@ def start_grpc_server(
     from .serve import ensure_prometheus_dir
     from .serve import PROMETHEUS_SERVER_NAME
     from ._internal.utils import reserve_free_port
-    from ._internal.resource import CpuResource
     from ._internal.utils.circus import create_standalone_arbiter
     from ._internal.utils.analytics import track_serve
 
@@ -304,7 +301,7 @@ def start_grpc_server(
                 args=args,
                 use_sockets=False,
                 working_dir=working_dir,
-                numprocesses=api_workers or math.ceil(CpuResource.from_system()),
+                numprocesses=api_workers,
             )
         )
 
diff --git a/pyproject.toml b/pyproject.toml
index 8a938ffc292..09c108b8896 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -379,7 +379,7 @@ exclude = [
   "bazel-*",
 ]
 analysis.useLibraryCodeForTypes = true
-analysis.stubPath = "./typings/"
+analysis.stubPath = "typings/"
 strictListInference = true
 strictDictionaryInference = true
 strictSetInference = true