diff --git a/pdm.lock b/pdm.lock
index eea1f7b2dba..51bda474a8e 100644
--- a/pdm.lock
+++ b/pdm.lock
@@ -5,7 +5,7 @@
 groups = ["default", "all", "aws", "docs", "grpc", "grpc-channelz", "grpc-reflection", "io", "io-image", "io-pandas", "monitor-otlp", "testing", "tooling", "tracing", "tracing-jaeger", "tracing-otlp", "tracing-zipkin"]
 strategy = ["inherit_metadata"]
 lock_version = "4.5.0"
-content_hash = "sha256:f6b631c837f69b66563c61b24913bf86075a0d2a05f1182d051721d5aad51d9a"
+content_hash = "sha256:c7d8a9bb26fc3e0842b297b12f3a0953494c2b7ed31431ae7ebc50a4edb2153d"
 
 [[metadata.targets]]
 requires_python = ">=3.9"
@@ -1836,12 +1836,12 @@ files = [
 
 [[package]]
 name = "nvidia-ml-py"
-version = "11.525.150"
+version = "12.560.30"
 summary = "Python Bindings for the NVIDIA Management Library"
 groups = ["default"]
 files = [
-    {file = "nvidia-ml-py-11.525.150.tar.gz", hash = "sha256:50af55b99ea167781102345a7de29bac94a57c8f38de6757ef9f945dd137c90a"},
-    {file = "nvidia_ml_py-11.525.150-py3-none-any.whl", hash = "sha256:a7c410f4a63a78119d8e9d969dbd22d3b86d8d39aa9316be56c2e6ba9271a5c3"},
+    {file = "nvidia-ml-py-12.560.30.tar.gz", hash = "sha256:f0254dc7400647680a072ee02509bfd46102b60bdfeca321576d4d4817e7fe97"},
+    {file = "nvidia_ml_py-12.560.30-py3-none-any.whl", hash = "sha256:fea371c94d63e38a611c17bbb85fe400e9c8ddb9e8684a9cd0e47786a4bc3c73"},
 ]
 
 [[package]]
diff --git a/pyproject.toml b/pyproject.toml
index f37d897f529..10e4009987d 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -37,7 +37,7 @@ dependencies = [
     "httpx",
     "inflection",
     "numpy",
-    "nvidia-ml-py<12",
+    "nvidia-ml-py",
     # OpenTelemetry is the server dependencies, rather than SDK
     # Since there are discrepancies among API and instrumentation packages,
     # we should always pin the set version of Opentelemetry suite
diff --git a/src/bentoml/_internal/container/frontend/dockerfile/templates/base_alpine.j2 b/src/bentoml/_internal/container/frontend/dockerfile/templates/base_alpine.j2
index 83f130ae7aa..c6e85375057 100644
--- a/src/bentoml/_internal/container/frontend/dockerfile/templates/base_alpine.j2
+++ b/src/bentoml/_internal/container/frontend/dockerfile/templates/base_alpine.j2
@@ -10,7 +10,7 @@ ENV ENV /root/.bashrc
 
 ENV UV_SYSTEM_PYTHON=1
 RUN curl -LO https://astral.sh/uv/install.sh && \
-    sh install.sh && rm install.sh && mv $HOME/.cargo/bin/uv /usr/local/bin/
+    sh install.sh && rm install.sh && mv $HOME/.local/bin/uv /usr/local/bin/
 
 {% if __options__system_packages is not none %}
 # Install user-defined system package
diff --git a/src/bentoml/_internal/container/frontend/dockerfile/templates/base_amazonlinux.j2 b/src/bentoml/_internal/container/frontend/dockerfile/templates/base_amazonlinux.j2
index d7fd6647fd0..311a70d93fe 100644
--- a/src/bentoml/_internal/container/frontend/dockerfile/templates/base_amazonlinux.j2
+++ b/src/bentoml/_internal/container/frontend/dockerfile/templates/base_amazonlinux.j2
@@ -15,7 +15,7 @@ USER root
 
 ENV UV_SYSTEM_PYTHON=1
 RUN curl -LO https://astral.sh/uv/install.sh && \
-    sh install.sh && rm install.sh && mv $HOME/.cargo/bin/uv /usr/local/bin/
+    sh install.sh && rm install.sh && mv $HOME/.local/bin/uv /usr/local/bin/
 {% if __options__system_packages is not none %}
 # Install user-defined system package
 {% call common.RUN(__enable_buildkit__) -%} {{ common.mount_cache("/var/cache/yum") }} {% endcall -%} set -eux && \
diff --git a/src/bentoml/_internal/container/frontend/dockerfile/templates/base_debian.j2 b/src/bentoml/_internal/container/frontend/dockerfile/templates/base_debian.j2
index 66e86cb79c0..82f67cf6369 100644
--- a/src/bentoml/_internal/container/frontend/dockerfile/templates/base_debian.j2
+++ b/src/bentoml/_internal/container/frontend/dockerfile/templates/base_debian.j2
@@ -14,5 +14,5 @@ RUN rm -f /etc/apt/apt.conf.d/docker-clean; echo 'Binary::apt::APT::Keep-Downloa
 
 ENV UV_SYSTEM_PYTHON=1
 RUN curl -LO https://astral.sh/uv/install.sh && \
-    sh install.sh && rm install.sh && mv $HOME/.cargo/bin/uv /usr/local/bin/
+    sh install.sh && rm install.sh && mv $HOME/.local/bin/uv /usr/local/bin/
 {% endblock %}
diff --git a/src/bentoml/_internal/container/frontend/dockerfile/templates/base_ubi8.j2 b/src/bentoml/_internal/container/frontend/dockerfile/templates/base_ubi8.j2
index f01c242b967..dfe6773bb53 100644
--- a/src/bentoml/_internal/container/frontend/dockerfile/templates/base_ubi8.j2
+++ b/src/bentoml/_internal/container/frontend/dockerfile/templates/base_ubi8.j2
@@ -9,7 +9,7 @@ USER root
 
 ENV UV_SYSTEM_PYTHON=1
 RUN curl -LO https://astral.sh/uv/install.sh && \
-    sh install.sh && rm install.sh && mv $HOME/.cargo/bin/uv /usr/local/bin/
+    sh install.sh && rm install.sh && mv $HOME/.local/bin/uv /usr/local/bin/
 {% if __options__system_packages is not none %}
 # Install user-defined system package
 {% call common.RUN(__enable_buildkit__) -%} {{ common.mount_cache("/var/cache/yum") }} {% endcall -%} set -eux && \
diff --git a/src/bentoml/_internal/container/frontend/dockerfile/templates/base_v2.j2 b/src/bentoml/_internal/container/frontend/dockerfile/templates/base_v2.j2
index ec1936a5f4d..632ab2124d9 100644
--- a/src/bentoml/_internal/container/frontend/dockerfile/templates/base_v2.j2
+++ b/src/bentoml/_internal/container/frontend/dockerfile/templates/base_v2.j2
@@ -66,7 +66,7 @@ RUN {{ command }}
 {% endfor %}
 
 RUN curl -LO https://astral.sh/uv/install.sh && \
-    sh install.sh && rm install.sh && mv $HOME/.cargo/bin/uv /usr/local/bin/ && uv venv --python {{ __options__python_version }}
+    sh install.sh && rm install.sh && mv $HOME/.local/bin/uv /usr/local/bin/ && uv venv --python {{ __options__python_version }}
 {% set __pip_cache__ = common.mount_cache("/root/.cache/") %}
 {% if __pip_preheat_packages__ %}
 {% for value in __pip_preheat_packages__ -%}