From 5aa0c5a7aee416a7a711ec24bc083c2edf71ee40 Mon Sep 17 00:00:00 2001
From: Tushar Sharma <tusharma@nvidia.com>
Date: Mon, 3 Nov 2025 12:44:40 -0800
Subject: [PATCH 01/14] initial commit for frontend image deployment

Signed-off-by: Tushar Sharma <tusharma@nvidia.com>
---
 container/Dockerfile                     | 18 ++++++++++++++++++
 container/run.sh                         |  2 +-
 examples/backends/sglang/deploy/agg.yaml | 10 +++++-----
 3 files changed, 24 insertions(+), 6 deletions(-)

diff --git a/container/Dockerfile b/container/Dockerfile
index b17c57ca7b..0afa72173d 100644
--- a/container/Dockerfile
+++ b/container/Dockerfile
@@ -401,3 +401,21 @@ RUN --mount=type=bind,source=./container/launch_message.txt,target=/opt/dynamo/l
 
 ENTRYPOINT ["/opt/nvidia/nvidia_entrypoint.sh"]
 CMD []
+
+FROM nvcr.io/nvidia/distroless/python:3.12-v3.5.0 as frontend
+
+COPY --from=dev /opt/dynamo/wheelhouse/ /opt/dynamo/wheelhouse/
+
+RUN pip install /opt/dynamo/wheelhouse/ai_dynamo_runtime*.whl /opt/dynamo/wheelhouse/ai_dynamo*any.whl /opt/dynamo/wheelhouse/nixl/nixl*.whl
+RUN if [ "$ENABLE_KVBM" = "true" ]; then \
+        pip install /opt/dynamo/wheelhouse/kvbm*.whl; \
+    fi
+
+COPY benchmarks/ /opt/dynamo/benchmarks/
+RUN pip install /opt/dynamo/benchmarks/
+
+COPY launch/ /opt/dynamo/launch/
+COPY lib/ /opt/dynamo/lib/
+COPY components/ /opt/dynamo/components/
+
+CMD []
\ No newline at end of file
diff --git a/container/run.sh b/container/run.sh
index 2f16b3e729..2b2e62f0dd 100755
--- a/container/run.sh
+++ b/container/run.sh
@@ -42,7 +42,7 @@ ENVIRONMENT_VARIABLES=
 REMAINING_ARGS=
 INTERACTIVE=
 USE_NIXL_GDS=
-RUNTIME=nvidia
+RUNTIME=
 WORKDIR=/workspace
 NETWORK=host
 
diff --git a/examples/backends/sglang/deploy/agg.yaml b/examples/backends/sglang/deploy/agg.yaml
index c6302906b6..7f14bf937d 100644
--- a/examples/backends/sglang/deploy/agg.yaml
+++ b/examples/backends/sglang/deploy/agg.yaml
@@ -4,19 +4,19 @@
 apiVersion: nvidia.com/v1alpha1
 kind: DynamoGraphDeployment
 metadata:
-  name: sglang-agg
+  name: tusharma-sglang-agg
 spec:
   services:
     Frontend:
-      dynamoNamespace: sglang-agg
+      dynamoNamespace: tusharma
       componentType: frontend
       replicas: 1
       extraPodSpec:
         mainContainer:
-          image: my-registry/sglang-runtime:my-tag
+          image: nvcr.io/nvidian/dynamo-dev/tusharma-frontend:0.2.0
     decode:
       envFromSecret: hf-token-secret
-      dynamoNamespace: sglang-agg
+      dynamoNamespace: tusharma
       componentType: worker
       replicas: 1
       resources:
@@ -24,7 +24,7 @@ spec:
           gpu: "1"
       extraPodSpec:
         mainContainer:
-          image: my-registry/sglang-runtime:my-tag
+          image: nvcr.io/nvidia/ai-dynamo/sglang-runtime:0.6.0
           workingDir: /workspace/examples/backends/sglang
           command:
           - python3

From 48ced8fe20af83e4ff44c0ce1d5757b6d6af58f0 Mon Sep 17 00:00:00 2001
From: Tushar Sharma <tusharma@nvidia.com>
Date: Tue, 4 Nov 2025 10:37:54 -0800
Subject: [PATCH 02/14] fixes

Signed-off-by: Tushar Sharma <tusharma@nvidia.com>
---
 container/Dockerfile | 55 +++++++++++++++++++++++++++++++-------------
 1 file changed, 39 insertions(+), 16 deletions(-)

diff --git a/container/Dockerfile b/container/Dockerfile
index 0afa72173d..82bbd22fc5 100644
--- a/container/Dockerfile
+++ b/container/Dockerfile
@@ -7,12 +7,12 @@
 ##################################
 
 # Base image configuration
-ARG BASE_IMAGE="nvcr.io/nvidia/cuda-dl-base"
+ARG BASE_IMAGE="nvcr.io/nvidia/base/ubuntu"
 # TODO OPS-612: NCCL will hang with 25.03, so use 25.01 for now
 # Please check https://github.com/ai-dynamo/dynamo/pull/1065
 # for details and reproducer to manually test if the image
 # can be updated to later versions.
-ARG BASE_IMAGE_TAG="25.01-cuda12.8-devel-ubuntu24.04"
+ARG BASE_IMAGE_TAG="noble-20250619"
 
 # Build configuration
 ARG ENABLE_KVBM=false
@@ -322,9 +322,9 @@ RUN --mount=type=secret,id=aws-key-id,env=AWS_ACCESS_KEY_ID \
     fi && \
     /tmp/use-sccache.sh show-stats "Dynamo"
 
-##############################################
+###################################################
 ########## Dev entrypoint image ##############
-##############################################
+###################################################
 
 FROM base AS dev
 
@@ -333,7 +333,8 @@ ARG ARCH_ALT
 
 # Application environment variables
 ENV DYNAMO_HOME=/opt/dynamo \
-    CARGO_TARGET_DIR=/opt/dynamo/target
+    CARGO_TARGET_DIR=/opt/dynamo/target \
+    CARGO_HOME=/usr/local/cargo
 
 # NIXL environment variables
 ENV NIXL_PREFIX=/opt/nvidia/nvda_nixl \
@@ -402,20 +403,42 @@ RUN --mount=type=bind,source=./container/launch_message.txt,target=/opt/dynamo/l
 ENTRYPOINT ["/opt/nvidia/nvidia_entrypoint.sh"]
 CMD []
 
-FROM nvcr.io/nvidia/distroless/python:3.12-v3.5.0 as frontend
+FROM ${BASE_IMAGE}:${BASE_IMAGE_TAG} AS frontend
 
-COPY --from=dev /opt/dynamo/wheelhouse/ /opt/dynamo/wheelhouse/
+# Install uv package manager
+COPY --from=ghcr.io/astral-sh/uv:latest /uv /uvx /bin/
 
-RUN pip install /opt/dynamo/wheelhouse/ai_dynamo_runtime*.whl /opt/dynamo/wheelhouse/ai_dynamo*any.whl /opt/dynamo/wheelhouse/nixl/nixl*.whl
-RUN if [ "$ENABLE_KVBM" = "true" ]; then \
-        pip install /opt/dynamo/wheelhouse/kvbm*.whl; \
-    fi
+# Create and activate virtual environment
+ARG PYTHON_VERSION
+RUN mkdir -p /opt/dynamo/venv && \
+    uv venv /opt/dynamo/venv --python $PYTHON_VERSION
 
-COPY benchmarks/ /opt/dynamo/benchmarks/
-RUN pip install /opt/dynamo/benchmarks/
+ENV VIRTUAL_ENV=/opt/dynamo/venv \
+    PATH="/opt/dynamo/venv/bin:${PATH}"
 
-COPY launch/ /opt/dynamo/launch/
-COPY lib/ /opt/dynamo/lib/
-COPY components/ /opt/dynamo/components/
+COPY --from=dev /opt/dynamo/wheelhouse/ /opt/dynamo/wheelhouse/
+RUN uv pip install \
+    /opt/dynamo/wheelhouse/ai_dynamo_runtime*.whl \
+    /opt/dynamo/wheelhouse/ai_dynamo*any.whl \
+    /opt/dynamo/wheelhouse/nixl/nixl*.whl && \
+    if [ "$ENABLE_KVBM" = "true" ]; then \
+        uv pip install /opt/dynamo/wheelhouse/kvbm*.whl; \
+    fi \
+    && cd /opt/dynamo/benchmarks \
+    && UV_GIT_LFS=1 uv pip install --no-cache . \
+    && cd - \
+    && rm -rf /opt/dynamo/benchmarks
 
+RUN --mount=type=bind,source=./container/deps/requirements.txt,target=/tmp/requirements.txt \
+    UV_GIT_LFS=1 uv pip install \
+        --no-cache \
+        --requirement /tmp/requirements.txt
+
+# Setup launch banner
+RUN --mount=type=bind,source=./container/launch_message.txt,target=/opt/dynamo/launch_message.txt \
+    sed '/^#\s/d' /opt/dynamo/launch_message.txt > ~/.launch_screen && \
+    echo "cat ~/.launch_screen" >> ~/.bashrc && \
+    echo "source $VIRTUAL_ENV/bin/activate" >> ~/.bashrc
+
+ENTRYPOINT ["/opt/nvidia/nvidia_entrypoint.sh"]
 CMD []
\ No newline at end of file

From 55c140b0d99b1a356bc6fc95b2cb1ab4621ec91e Mon Sep 17 00:00:00 2001
From: Tushar Sharma <tusharma@nvidia.com>
Date: Wed, 5 Nov 2025 20:29:25 -0800
Subject: [PATCH 03/14] Initial commit for frontend image

Signed-off-by: Tushar Sharma <tusharma@nvidia.com>
---
 container/Dockerfile | 12 +++++++++++-
 1 file changed, 11 insertions(+), 1 deletion(-)

diff --git a/container/Dockerfile b/container/Dockerfile
index 82bbd22fc5..4ff237f2ff 100644
--- a/container/Dockerfile
+++ b/container/Dockerfile
@@ -408,6 +408,11 @@ FROM ${BASE_IMAGE}:${BASE_IMAGE_TAG} AS frontend
 # Install uv package manager
 COPY --from=ghcr.io/astral-sh/uv:latest /uv /uvx /bin/
 
+COPY --from=dev /usr/bin/nats-server /usr/bin/nats-server
+COPY --from=dev /usr/local/bin/etcd/ /usr/local/bin/etcd/
+# Add ETCD to PATH
+ENV PATH=/usr/local/bin/etcd:$PATH
+
 # Create and activate virtual environment
 ARG PYTHON_VERSION
 RUN mkdir -p /opt/dynamo/venv && \
@@ -416,6 +421,7 @@ RUN mkdir -p /opt/dynamo/venv && \
 ENV VIRTUAL_ENV=/opt/dynamo/venv \
     PATH="/opt/dynamo/venv/bin:${PATH}"
 
+COPY benchmarks/ /opt/dynamo/benchmarks/
 COPY --from=dev /opt/dynamo/wheelhouse/ /opt/dynamo/wheelhouse/
 RUN uv pip install \
     /opt/dynamo/wheelhouse/ai_dynamo_runtime*.whl \
@@ -430,9 +436,13 @@ RUN uv pip install \
     && rm -rf /opt/dynamo/benchmarks
 
 RUN --mount=type=bind,source=./container/deps/requirements.txt,target=/tmp/requirements.txt \
+    --mount=type=bind,source=./container/deps/requirements.test.txt,target=/tmp/requirements.test.txt \
     UV_GIT_LFS=1 uv pip install \
         --no-cache \
-        --requirement /tmp/requirements.txt
+        --requirement /tmp/requirements.txt \
+        --requirement /tmp/requirements.test.txt
+
+COPY . /workspace/
 
 # Setup launch banner
 RUN --mount=type=bind,source=./container/launch_message.txt,target=/opt/dynamo/launch_message.txt \

From 75b36ac1fda6d4e53fc4173b0dd56cbfcd77a343 Mon Sep 17 00:00:00 2001
From: Tushar Sharma <tusharma@nvidia.com>
Date: Tue, 11 Nov 2025 14:57:09 -0800
Subject: [PATCH 04/14] Change base image of dev stage

Signed-off-by: Tushar Sharma <tusharma@nvidia.com>
---
 container/Dockerfile | 60 ++++++--------------------------------------
 1 file changed, 7 insertions(+), 53 deletions(-)

diff --git a/container/Dockerfile b/container/Dockerfile
index 4ff237f2ff..8e5d191a80 100644
--- a/container/Dockerfile
+++ b/container/Dockerfile
@@ -323,10 +323,10 @@ RUN --mount=type=secret,id=aws-key-id,env=AWS_ACCESS_KEY_ID \
     /tmp/use-sccache.sh show-stats "Dynamo"
 
 ###################################################
-########## Dev entrypoint image ##############
+########## Frontend entrypoint image ##############
 ###################################################
 
-FROM base AS dev
+FROM nvcr.io/nvidia/base/ubuntu:noble-20250619 AS frontend
 
 ARG ENABLE_KVBM
 ARG ARCH_ALT
@@ -356,6 +356,7 @@ COPY --from=wheel_builder $CARGO_HOME $CARGO_HOME
 RUN apt-get update -y \
     && apt-get install -y --no-install-recommends \
         # required for AIC perf files
+        ca-certificates \
         git \
         git-lfs \
         # rust build packages
@@ -365,6 +366,9 @@ RUN apt-get update -y \
     && apt-get clean \
     && rm -rf /var/lib/apt/lists/*
 
+# Install uv package manager
+COPY --from=ghcr.io/astral-sh/uv:latest /uv /uvx /bin/
+
 # Create and activate virtual environment
 ARG PYTHON_VERSION
 RUN mkdir -p /opt/dynamo/venv && \
@@ -400,55 +404,5 @@ RUN --mount=type=bind,source=./container/launch_message.txt,target=/opt/dynamo/l
     echo "cat ~/.launch_screen" >> ~/.bashrc && \
     echo "source $VIRTUAL_ENV/bin/activate" >> ~/.bashrc
 
-ENTRYPOINT ["/opt/nvidia/nvidia_entrypoint.sh"]
-CMD []
-
-FROM ${BASE_IMAGE}:${BASE_IMAGE_TAG} AS frontend
-
-# Install uv package manager
-COPY --from=ghcr.io/astral-sh/uv:latest /uv /uvx /bin/
-
-COPY --from=dev /usr/bin/nats-server /usr/bin/nats-server
-COPY --from=dev /usr/local/bin/etcd/ /usr/local/bin/etcd/
-# Add ETCD to PATH
-ENV PATH=/usr/local/bin/etcd:$PATH
-
-# Create and activate virtual environment
-ARG PYTHON_VERSION
-RUN mkdir -p /opt/dynamo/venv && \
-    uv venv /opt/dynamo/venv --python $PYTHON_VERSION
-
-ENV VIRTUAL_ENV=/opt/dynamo/venv \
-    PATH="/opt/dynamo/venv/bin:${PATH}"
-
-COPY benchmarks/ /opt/dynamo/benchmarks/
-COPY --from=dev /opt/dynamo/wheelhouse/ /opt/dynamo/wheelhouse/
-RUN uv pip install \
-    /opt/dynamo/wheelhouse/ai_dynamo_runtime*.whl \
-    /opt/dynamo/wheelhouse/ai_dynamo*any.whl \
-    /opt/dynamo/wheelhouse/nixl/nixl*.whl && \
-    if [ "$ENABLE_KVBM" = "true" ]; then \
-        uv pip install /opt/dynamo/wheelhouse/kvbm*.whl; \
-    fi \
-    && cd /opt/dynamo/benchmarks \
-    && UV_GIT_LFS=1 uv pip install --no-cache . \
-    && cd - \
-    && rm -rf /opt/dynamo/benchmarks
-
-RUN --mount=type=bind,source=./container/deps/requirements.txt,target=/tmp/requirements.txt \
-    --mount=type=bind,source=./container/deps/requirements.test.txt,target=/tmp/requirements.test.txt \
-    UV_GIT_LFS=1 uv pip install \
-        --no-cache \
-        --requirement /tmp/requirements.txt \
-        --requirement /tmp/requirements.test.txt
-
-COPY . /workspace/
-
-# Setup launch banner
-RUN --mount=type=bind,source=./container/launch_message.txt,target=/opt/dynamo/launch_message.txt \
-    sed '/^#\s/d' /opt/dynamo/launch_message.txt > ~/.launch_screen && \
-    echo "cat ~/.launch_screen" >> ~/.bashrc && \
-    echo "source $VIRTUAL_ENV/bin/activate" >> ~/.bashrc
-
-ENTRYPOINT ["/opt/nvidia/nvidia_entrypoint.sh"]
+ENTRYPOINT ["/bin/bash", "-c"]
 CMD []
\ No newline at end of file

From a48f140971a69a61d43766e83280631ba7f3f8d3 Mon Sep 17 00:00:00 2001
From: Tushar Sharma <tusharma@nvidia.com>
Date: Tue, 11 Nov 2025 15:02:15 -0800
Subject: [PATCH 05/14] revert some unncessary changes that were brought in

Signed-off-by: Tushar Sharma <tusharma@nvidia.com>
---
 container/Dockerfile                     |  4 ++--
 container/run.sh                         |  2 +-
 examples/backends/sglang/deploy/agg.yaml | 10 +++++-----
 3 files changed, 8 insertions(+), 8 deletions(-)

diff --git a/container/Dockerfile b/container/Dockerfile
index 8e5d191a80..0773d3a6ad 100644
--- a/container/Dockerfile
+++ b/container/Dockerfile
@@ -7,12 +7,12 @@
 ##################################
 
 # Base image configuration
-ARG BASE_IMAGE="nvcr.io/nvidia/base/ubuntu"
+ARG BASE_IMAGE="nvcr.io/nvidia/cuda-dl-base"
 # TODO OPS-612: NCCL will hang with 25.03, so use 25.01 for now
 # Please check https://github.com/ai-dynamo/dynamo/pull/1065
 # for details and reproducer to manually test if the image
 # can be updated to later versions.
-ARG BASE_IMAGE_TAG="noble-20250619"
+ARG BASE_IMAGE_TAG="25.01-cuda12.8-devel-ubuntu24.04"
 
 # Build configuration
 ARG ENABLE_KVBM=false
diff --git a/container/run.sh b/container/run.sh
index 2b2e62f0dd..2f16b3e729 100755
--- a/container/run.sh
+++ b/container/run.sh
@@ -42,7 +42,7 @@ ENVIRONMENT_VARIABLES=
 REMAINING_ARGS=
 INTERACTIVE=
 USE_NIXL_GDS=
-RUNTIME=
+RUNTIME=nvidia
 WORKDIR=/workspace
 NETWORK=host
 
diff --git a/examples/backends/sglang/deploy/agg.yaml b/examples/backends/sglang/deploy/agg.yaml
index 7f14bf937d..c6302906b6 100644
--- a/examples/backends/sglang/deploy/agg.yaml
+++ b/examples/backends/sglang/deploy/agg.yaml
@@ -4,19 +4,19 @@
 apiVersion: nvidia.com/v1alpha1
 kind: DynamoGraphDeployment
 metadata:
-  name: tusharma-sglang-agg
+  name: sglang-agg
 spec:
   services:
     Frontend:
-      dynamoNamespace: tusharma
+      dynamoNamespace: sglang-agg
       componentType: frontend
       replicas: 1
       extraPodSpec:
         mainContainer:
-          image: nvcr.io/nvidian/dynamo-dev/tusharma-frontend:0.2.0
+          image: my-registry/sglang-runtime:my-tag
     decode:
       envFromSecret: hf-token-secret
-      dynamoNamespace: tusharma
+      dynamoNamespace: sglang-agg
       componentType: worker
       replicas: 1
       resources:
@@ -24,7 +24,7 @@ spec:
           gpu: "1"
       extraPodSpec:
         mainContainer:
-          image: nvcr.io/nvidia/ai-dynamo/sglang-runtime:0.6.0
+          image: my-registry/sglang-runtime:my-tag
           workingDir: /workspace/examples/backends/sglang
           command:
           - python3

From 4d932ae6af0ab9b621d0c77e2d5312dbe41e1cad Mon Sep 17 00:00:00 2001
From: Tushar Sharma <tusharma@nvidia.com>
Date: Tue, 11 Nov 2025 19:54:57 -0800
Subject: [PATCH 06/14] commit latest changes from dockerfile

Signed-off-by: Tushar Sharma <tusharma@nvidia.com>
---
 .../workflows/container-validation-dynamo.yml |  2 +-
 container/Dockerfile                          | 33 +++++++++++++------
 2 files changed, 24 insertions(+), 11 deletions(-)

diff --git a/.github/workflows/container-validation-dynamo.yml b/.github/workflows/container-validation-dynamo.yml
index f982a2d08a..0e638a17fd 100644
--- a/.github/workflows/container-validation-dynamo.yml
+++ b/.github/workflows/container-validation-dynamo.yml
@@ -42,7 +42,7 @@ jobs:
         env:
           GITHUB_TOKEN: ${{ secrets.CI_TOKEN }}
         run: |
-          ./container/build.sh --tag ${{ steps.define_image_tag.outputs.image_tag }} --target dev --framework none
+          ./container/build.sh --tag ${{ steps.define_image_tag.outputs.image_tag }} --target frontend --framework none
       - name: Start services with docker-compose
         working-directory: ./deploy
         run: |
diff --git a/container/Dockerfile b/container/Dockerfile
index 7a81b194fe..ed790f6799 100644
--- a/container/Dockerfile
+++ b/container/Dockerfile
@@ -322,9 +322,14 @@ RUN --mount=type=secret,id=aws-key-id,env=AWS_ACCESS_KEY_ID \
     fi && \
     /tmp/use-sccache.sh show-stats "Dynamo"
 
+
 ###################################################
 ########## Frontend entrypoint image ##############
 ###################################################
+# This is the main frontend image that includes:
+#   - Python environment with Dynamo packages
+#   - Rust toolchain for development
+#   - NATS and ETCD servers
 
 FROM nvcr.io/nvidia/base/ubuntu:noble-20250619 AS frontend
 
@@ -334,7 +339,9 @@ ARG ARCH_ALT
 # Application environment variables
 ENV DYNAMO_HOME=/opt/dynamo \
     CARGO_TARGET_DIR=/opt/dynamo/target \
-    CARGO_HOME=/usr/local/cargo
+    RUSTUP_HOME=/usr/local/rustup \
+    CARGO_HOME=/usr/local/cargo \
+    PATH=/usr/local/cargo/bin:$PATH
 
 RUN apt-get update -y \
     && apt-get install -y --no-install-recommends \
@@ -342,23 +349,24 @@ RUN apt-get update -y \
         ca-certificates \
         git \
         git-lfs \
+        # curl for downloading files
+        curl \
         # rust build packages
+        build-essential \
+        gcc \
+        g++ \
         clang \
         libclang-dev \
         protobuf-compiler \
-        # sudo for dev stage
-        sudo \
+        # libstdc++6 is needed for EPP runtime (included with g++)
     && apt-get clean \
-    && rm -rf /var/lib/apt/lists/* \
-    # Add sudo privileges to dynamo user
-    && echo "dynamo ALL=(ALL) NOPASSWD:ALL" > /etc/sudoers.d/dynamo \
-    && chmod 0440 /etc/sudoers.d/dynamo
+    && rm -rf /var/lib/apt/lists/*
 
 # Create dynamo user with group 0 for OpenShift compatibility
 RUN userdel -r ubuntu > /dev/null 2>&1 || true \
     && useradd -m -s /bin/bash -g 0 dynamo \
     && [ `id -u dynamo` -eq 1000 ] \
-    && mkdir -p /home/dynamo/.cache \
+    && mkdir -p /home/dynamo/.cache /opt/dynamo /workspace \
     && chown -R dynamo: /opt/dynamo /home/dynamo /workspace \
     && chmod -R g+w /opt/dynamo /home/dynamo/.cache /workspace
 
@@ -390,6 +398,11 @@ ENV NIXL_PREFIX=/opt/nvidia/nvda_nixl \
     NIXL_PLUGIN_DIR=/opt/nvidia/nvda_nixl/lib/${ARCH_ALT}-linux-gnu/plugins
 ENV LD_LIBRARY_PATH=${NIXL_LIB_DIR}:${NIXL_PLUGIN_DIR}:/usr/local/ucx/lib:/usr/local/ucx/lib/ucx:${LD_LIBRARY_PATH}
 
+# Copy NATS and ETCD from base stage
+COPY --chown=dynamo: --from=base /usr/bin/nats-server /usr/bin/nats-server
+COPY --chown=dynamo: --from=base /usr/local/bin/etcd/ /usr/local/bin/etcd/
+ENV PATH=/usr/local/bin/etcd/:$PATH
+
 # Copy ucx and nixl libs
 COPY --chown=dynamo: --from=wheel_builder /usr/local/ucx/ /usr/local/ucx/
 COPY --chown=dynamo: --from=wheel_builder ${NIXL_PREFIX}/ ${NIXL_PREFIX}/
@@ -399,6 +412,7 @@ COPY --chown=dynamo: --from=wheel_builder /opt/nvidia/nvda_nixl/lib64/. ${NIXL_L
 COPY --chown=dynamo: --from=wheel_builder /opt/dynamo/dist/nixl/ /opt/dynamo/wheelhouse/nixl/
 COPY --chown=dynamo: --from=wheel_builder /opt/dynamo/dist/*.whl /opt/dynamo/wheelhouse/
 COPY --chown=dynamo: --from=wheel_builder $CARGO_TARGET_DIR $CARGO_TARGET_DIR
+COPY --chown=dynamo: --from=wheel_builder $RUSTUP_HOME $RUSTUP_HOME
 COPY --chown=dynamo: --from=wheel_builder $CARGO_HOME $CARGO_HOME
 
 COPY --chown=dynamo: ./ /workspace/
@@ -425,5 +439,4 @@ RUN chmod 755 /opt/dynamo/.launch_screen && \
 
 USER dynamo
 
-ENTRYPOINT ["/bin/bash", "-c"]
-CMD []
\ No newline at end of file
+CMD ["/bin/bash"]
\ No newline at end of file

From 560e5bd4733bae43efd7d9b4d9b34089889040c6 Mon Sep 17 00:00:00 2001
From: Tushar Sharma <tusharma@nvidia.com>
Date: Tue, 11 Nov 2025 21:38:33 -0800
Subject: [PATCH 07/14] Add frontend container to seperately build frontend
 image with epp

Signed-off-by: Tushar Sharma <tusharma@nvidia.com>
---
 container/Dockerfile          | 46 +++++-----------
 container/Dockerfile.frontend | 98 +++++++++++++++++++++++++++++++++++
 2 files changed, 112 insertions(+), 32 deletions(-)
 create mode 100644 container/Dockerfile.frontend

diff --git a/container/Dockerfile b/container/Dockerfile
index ed790f6799..4d3a513004 100644
--- a/container/Dockerfile
+++ b/container/Dockerfile
@@ -323,50 +323,40 @@ RUN --mount=type=secret,id=aws-key-id,env=AWS_ACCESS_KEY_ID \
     /tmp/use-sccache.sh show-stats "Dynamo"
 
 
-###################################################
-########## Frontend entrypoint image ##############
-###################################################
-# This is the main frontend image that includes:
-#   - Python environment with Dynamo packages
-#   - Rust toolchain for development
-#   - NATS and ETCD servers
-
-FROM nvcr.io/nvidia/base/ubuntu:noble-20250619 AS frontend
+##############################################
+########## Dev entrypoint image ##############
+##############################################
+FROM base AS dev
 
 ARG ENABLE_KVBM
 ARG ARCH_ALT
 
 # Application environment variables
 ENV DYNAMO_HOME=/opt/dynamo \
-    CARGO_TARGET_DIR=/opt/dynamo/target \
-    RUSTUP_HOME=/usr/local/rustup \
-    CARGO_HOME=/usr/local/cargo \
-    PATH=/usr/local/cargo/bin:$PATH
+    CARGO_TARGET_DIR=/opt/dynamo/target
 
 RUN apt-get update -y \
     && apt-get install -y --no-install-recommends \
         # required for AIC perf files
-        ca-certificates \
         git \
         git-lfs \
-        # curl for downloading files
-        curl \
         # rust build packages
-        build-essential \
-        gcc \
-        g++ \
         clang \
         libclang-dev \
         protobuf-compiler \
-        # libstdc++6 is needed for EPP runtime (included with g++)
+        # sudo for dev stage
+        sudo \
     && apt-get clean \
-    && rm -rf /var/lib/apt/lists/*
+    && rm -rf /var/lib/apt/lists/* \
+    # Add sudo privileges to dynamo user
+    && echo "dynamo ALL=(ALL) NOPASSWD:ALL" > /etc/sudoers.d/dynamo \
+    && chmod 0440 /etc/sudoers.d/dynamo
 
 # Create dynamo user with group 0 for OpenShift compatibility
 RUN userdel -r ubuntu > /dev/null 2>&1 || true \
     && useradd -m -s /bin/bash -g 0 dynamo \
     && [ `id -u dynamo` -eq 1000 ] \
-    && mkdir -p /home/dynamo/.cache /opt/dynamo /workspace \
+    && mkdir -p /home/dynamo/.cache \
     && chown -R dynamo: /opt/dynamo /home/dynamo /workspace \
     && chmod -R g+w /opt/dynamo /home/dynamo/.cache /workspace
 
@@ -374,9 +364,6 @@ RUN userdel -r ubuntu > /dev/null 2>&1 || true \
 USER dynamo
 ENV HOME=/home/dynamo
 
-# Install uv package manager
-COPY --from=ghcr.io/astral-sh/uv:latest /uv /uvx /bin/
-
 # Create and activate virtual environment
 ARG PYTHON_VERSION
 RUN uv venv /opt/dynamo/venv --python $PYTHON_VERSION
@@ -398,11 +385,6 @@ ENV NIXL_PREFIX=/opt/nvidia/nvda_nixl \
     NIXL_PLUGIN_DIR=/opt/nvidia/nvda_nixl/lib/${ARCH_ALT}-linux-gnu/plugins
 ENV LD_LIBRARY_PATH=${NIXL_LIB_DIR}:${NIXL_PLUGIN_DIR}:/usr/local/ucx/lib:/usr/local/ucx/lib/ucx:${LD_LIBRARY_PATH}
 
-# Copy NATS and ETCD from base stage
-COPY --chown=dynamo: --from=base /usr/bin/nats-server /usr/bin/nats-server
-COPY --chown=dynamo: --from=base /usr/local/bin/etcd/ /usr/local/bin/etcd/
-ENV PATH=/usr/local/bin/etcd/:$PATH
-
 # Copy ucx and nixl libs
 COPY --chown=dynamo: --from=wheel_builder /usr/local/ucx/ /usr/local/ucx/
 COPY --chown=dynamo: --from=wheel_builder ${NIXL_PREFIX}/ ${NIXL_PREFIX}/
@@ -412,7 +394,6 @@ COPY --chown=dynamo: --from=wheel_builder /opt/nvidia/nvda_nixl/lib64/. ${NIXL_L
 COPY --chown=dynamo: --from=wheel_builder /opt/dynamo/dist/nixl/ /opt/dynamo/wheelhouse/nixl/
 COPY --chown=dynamo: --from=wheel_builder /opt/dynamo/dist/*.whl /opt/dynamo/wheelhouse/
 COPY --chown=dynamo: --from=wheel_builder $CARGO_TARGET_DIR $CARGO_TARGET_DIR
-COPY --chown=dynamo: --from=wheel_builder $RUSTUP_HOME $RUSTUP_HOME
 COPY --chown=dynamo: --from=wheel_builder $CARGO_HOME $CARGO_HOME
 
 COPY --chown=dynamo: ./ /workspace/
@@ -439,4 +420,5 @@ RUN chmod 755 /opt/dynamo/.launch_screen && \
 
 USER dynamo
 
-CMD ["/bin/bash"]
\ No newline at end of file
+ENTRYPOINT ["/opt/nvidia/nvidia_entrypoint.sh"]
+CMD []
\ No newline at end of file
diff --git a/container/Dockerfile.frontend b/container/Dockerfile.frontend
new file mode 100644
index 0000000000..7ed8368a13
--- /dev/null
+++ b/container/Dockerfile.frontend
@@ -0,0 +1,98 @@
+# syntax=docker/dockerfile:1.10.0
+# SPDX-FileCopyrightText: Copyright (c) 2025 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
+# SPDX-License-Identifier: Apache-2.0
+
+ARG DYNAMO_BASE_IMAGE="dynamo:latest-none"
+ARG EPP_IMAGE="us-central1-docker.pkg.dev/k8s-staging-images/gateway-api-inference-extension/epp:v0.5.1-dirty"
+ARG PYTHON_VERSION=3.12
+ARG ENABLE_KVBM=false
+
+FROM ${DYNAMO_BASE_IMAGE} AS dynamo_base
+
+FROM ${EPP_IMAGE} AS epp
+
+FROM nvcr.io/nvidia/base/ubuntu:noble-20250619 AS frontend
+
+RUN apt-get update -y \
+    && apt-get install -y --no-install-recommends \
+        # required for AIC perf files
+        ca-certificates \
+        git \
+        git-lfs \
+        # curl for downloading files
+        curl \
+        libstdc++6 \
+    && apt-get clean \
+    && rm -rf /var/lib/apt/lists/*
+
+WORKDIR /
+COPY --from=epp /epp /epp
+
+# Create dynamo user with group 0 for OpenShift compatibility
+RUN userdel -r ubuntu > /dev/null 2>&1 || true \
+    && useradd -m -s /bin/bash -g 0 dynamo \
+    && [ `id -u dynamo` -eq 1000 ] \
+    && mkdir -p /home/dynamo/.cache /opt/dynamo \
+    && chown -R dynamo: /opt/dynamo /home/dynamo/.cache \
+    && chmod -R g+w /opt/dynamo /home/dynamo/.cache
+
+# Set HOME so ModelExpress can find the cache directory
+ENV HOME=/home/dynamo
+
+# Switch to dynamo user
+USER dynamo
+ENV DYNAMO_HOME=/opt/dynamo
+
+# Copy NATS and ETCD from base stage
+COPY --from=dynamo_base /usr/bin/nats-server /usr/bin/nats-server
+COPY --from=dynamo_base /usr/local/bin/etcd/ /usr/local/bin/etcd/
+ENV PATH=/usr/local/bin/etcd/:$PATH
+
+# Install uv package manager
+COPY --from=ghcr.io/astral-sh/uv:latest /uv /uvx /bin/
+
+# Create and activate virtual environment
+ARG PYTHON_VERSION
+RUN uv venv /opt/dynamo/venv --python $PYTHON_VERSION
+
+ENV VIRTUAL_ENV=/opt/dynamo/venv \
+    PATH="/opt/dynamo/venv/bin:${PATH}"
+
+# Copy built artifacts and workspace from dynamo_base (dev image)
+COPY --chown=dynamo: --from=dynamo_base /opt/dynamo/wheelhouse/ /opt/dynamo/wheelhouse/
+COPY --chown=dynamo: --from=dynamo_base /workspace/ /workspace/
+
+ARG ENABLE_KVBM
+RUN uv pip install \
+    /opt/dynamo/wheelhouse/ai_dynamo_runtime*.whl \
+    /opt/dynamo/wheelhouse/ai_dynamo*any.whl \
+    /opt/dynamo/wheelhouse/nixl/nixl*.whl && \
+    if [ "$ENABLE_KVBM" = "true" ]; then \
+        uv pip install /opt/dynamo/wheelhouse/kvbm*.whl; \
+    fi && \
+    cd /workspace/benchmarks && \
+    UV_GIT_LFS=1 uv pip install --no-cache .
+    
+# Install common and test dependencies
+RUN --mount=type=bind,source=./container/deps/requirements.txt,target=/tmp/requirements.txt \
+    --mount=type=bind,source=./container/deps/requirements.test.txt,target=/tmp/requirements.test.txt \
+    UV_GIT_LFS=1 uv pip install \
+        --no-cache \
+        --requirement /tmp/requirements.txt \
+        --requirement /tmp/requirements.test.txt
+
+
+# Setup launch banner in common directory accessible to all users
+RUN --mount=type=bind,source=./container/launch_message.txt,target=/opt/dynamo/launch_message.txt \
+    sed '/^#\s/d' /opt/dynamo/launch_message.txt > /opt/dynamo/.launch_screen
+
+# Setup environment for all users
+USER root
+RUN chmod 755 /opt/dynamo/.launch_screen && \
+    echo 'source /opt/dynamo/venv/bin/activate' >> /etc/bash.bashrc && \
+    echo 'cat /opt/dynamo/.launch_screen' >> /etc/bash.bashrc
+
+USER dynamo
+
+CMD ["/bin/bash"]
+

From b0a137d04385daf65a06bc992d27304fb9ef9ea0 Mon Sep 17 00:00:00 2001
From: Tushar Sharma <tusharma@nvidia.com>
Date: Tue, 11 Nov 2025 21:41:46 -0800
Subject: [PATCH 08/14] fix spacing in base dockerfile

Signed-off-by: Tushar Sharma <tusharma@nvidia.com>
---
 container/Dockerfile | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/container/Dockerfile b/container/Dockerfile
index 4d3a513004..4623b4d84a 100644
--- a/container/Dockerfile
+++ b/container/Dockerfile
@@ -322,10 +322,10 @@ RUN --mount=type=secret,id=aws-key-id,env=AWS_ACCESS_KEY_ID \
     fi && \
     /tmp/use-sccache.sh show-stats "Dynamo"
 
-
 ##############################################
 ########## Dev entrypoint image ##############
 ##############################################
+
 FROM base AS dev
 
 ARG ENABLE_KVBM
@@ -421,4 +421,4 @@ RUN chmod 755 /opt/dynamo/.launch_screen && \
 USER dynamo
 
 ENTRYPOINT ["/opt/nvidia/nvidia_entrypoint.sh"]
-CMD []
\ No newline at end of file
+CMD []

From 6573cc16ed600e14c71943a74c722f8c062e7891 Mon Sep 17 00:00:00 2001
From: Tushar Sharma <tusharma@nvidia.com>
Date: Tue, 11 Nov 2025 21:43:06 -0800
Subject: [PATCH 09/14] remove frontend from GHA workflow

Signed-off-by: Tushar Sharma <tusharma@nvidia.com>
---
 .github/workflows/container-validation-dynamo.yml | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/.github/workflows/container-validation-dynamo.yml b/.github/workflows/container-validation-dynamo.yml
index 0e638a17fd..f982a2d08a 100644
--- a/.github/workflows/container-validation-dynamo.yml
+++ b/.github/workflows/container-validation-dynamo.yml
@@ -42,7 +42,7 @@ jobs:
         env:
           GITHUB_TOKEN: ${{ secrets.CI_TOKEN }}
         run: |
-          ./container/build.sh --tag ${{ steps.define_image_tag.outputs.image_tag }} --target frontend --framework none
+          ./container/build.sh --tag ${{ steps.define_image_tag.outputs.image_tag }} --target dev --framework none
       - name: Start services with docker-compose
         working-directory: ./deploy
         run: |

From ac44519e2c62470e891632739cef5e980a794128 Mon Sep 17 00:00:00 2001
From: Tushar Sharma <tusharma@nvidia.com>
Date: Tue, 11 Nov 2025 21:50:49 -0800
Subject: [PATCH 10/14] update user permissions for frontend container

Signed-off-by: Tushar Sharma <tusharma@nvidia.com>
---
 container/Dockerfile.frontend | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/container/Dockerfile.frontend b/container/Dockerfile.frontend
index 7ed8368a13..42e1d98c3f 100644
--- a/container/Dockerfile.frontend
+++ b/container/Dockerfile.frontend
@@ -25,8 +25,6 @@ RUN apt-get update -y \
     && apt-get clean \
     && rm -rf /var/lib/apt/lists/*
 
-WORKDIR /
-COPY --from=epp /epp /epp
 
 # Create dynamo user with group 0 for OpenShift compatibility
 RUN userdel -r ubuntu > /dev/null 2>&1 || true \
@@ -43,6 +41,9 @@ ENV HOME=/home/dynamo
 USER dynamo
 ENV DYNAMO_HOME=/opt/dynamo
 
+WORKDIR /
+COPY --chown=dynamo: --from=epp /epp /epp
+
 # Copy NATS and ETCD from base stage
 COPY --from=dynamo_base /usr/bin/nats-server /usr/bin/nats-server
 COPY --from=dynamo_base /usr/local/bin/etcd/ /usr/local/bin/etcd/
@@ -61,7 +62,6 @@ ENV VIRTUAL_ENV=/opt/dynamo/venv \
 # Copy built artifacts and workspace from dynamo_base (dev image)
 COPY --chown=dynamo: --from=dynamo_base /opt/dynamo/wheelhouse/ /opt/dynamo/wheelhouse/
 COPY --chown=dynamo: --from=dynamo_base /workspace/ /workspace/
-
 ARG ENABLE_KVBM
 RUN uv pip install \
     /opt/dynamo/wheelhouse/ai_dynamo_runtime*.whl \

From c0c1f3d991d2e1f6846ff20d514ab169f9650702 Mon Sep 17 00:00:00 2001
From: Tushar Sharma <tusharma@nvidia.com>
Date: Wed, 12 Nov 2025 09:52:07 -0800
Subject: [PATCH 11/14] add documentation + productionize image

Signed-off-by: Tushar Sharma <tusharma@nvidia.com>
---
 container/Dockerfile.frontend | 21 ++++++++++-----
 container/README.md           | 51 +++++++++++++++++++++++++++++++++++
 2 files changed, 66 insertions(+), 6 deletions(-)

diff --git a/container/Dockerfile.frontend b/container/Dockerfile.frontend
index 42e1d98c3f..cd538b5973 100644
--- a/container/Dockerfile.frontend
+++ b/container/Dockerfile.frontend
@@ -30,9 +30,9 @@ RUN apt-get update -y \
 RUN userdel -r ubuntu > /dev/null 2>&1 || true \
     && useradd -m -s /bin/bash -g 0 dynamo \
     && [ `id -u dynamo` -eq 1000 ] \
-    && mkdir -p /home/dynamo/.cache /opt/dynamo \
-    && chown -R dynamo: /opt/dynamo /home/dynamo/.cache \
-    && chmod -R g+w /opt/dynamo /home/dynamo/.cache
+    && mkdir -p /home/dynamo/.cache /opt/dynamo /workspace \
+    && chown -R dynamo: /opt/dynamo /home/dynamo/.cache /workspace \
+    && chmod -R g+w /opt/dynamo /home/dynamo/.cache /workspace
 
 # Set HOME so ModelExpress can find the cache directory
 ENV HOME=/home/dynamo
@@ -59,10 +59,10 @@ RUN uv venv /opt/dynamo/venv --python $PYTHON_VERSION
 ENV VIRTUAL_ENV=/opt/dynamo/venv \
     PATH="/opt/dynamo/venv/bin:${PATH}"
 
+ARG ENABLE_KVBM
 # Copy built artifacts and workspace from dynamo_base (dev image)
+COPY --chown=dynamo: benchmarks/ /opt/dynamo/benchmarks/
 COPY --chown=dynamo: --from=dynamo_base /opt/dynamo/wheelhouse/ /opt/dynamo/wheelhouse/
-COPY --chown=dynamo: --from=dynamo_base /workspace/ /workspace/
-ARG ENABLE_KVBM
 RUN uv pip install \
     /opt/dynamo/wheelhouse/ai_dynamo_runtime*.whl \
     /opt/dynamo/wheelhouse/ai_dynamo*any.whl \
@@ -70,7 +70,7 @@ RUN uv pip install \
     if [ "$ENABLE_KVBM" = "true" ]; then \
         uv pip install /opt/dynamo/wheelhouse/kvbm*.whl; \
     fi && \
-    cd /workspace/benchmarks && \
+    cd /opt/dynamo/benchmarks && \
     UV_GIT_LFS=1 uv pip install --no-cache .
     
 # Install common and test dependencies
@@ -81,6 +81,15 @@ RUN --mount=type=bind,source=./container/deps/requirements.txt,target=/tmp/requi
         --requirement /tmp/requirements.txt \
         --requirement /tmp/requirements.test.txt
 
+# Copy tests, benchmarks, deploy and components for CI with correct ownership
+COPY --chown=dynamo: tests /workspace/tests
+COPY --chown=dynamo: examples /workspace/examples
+COPY --chown=dynamo: benchmarks /workspace/benchmarks
+COPY --chown=dynamo: deploy /workspace/deploy
+COPY --chown=dynamo: components/ /workspace/components/
+COPY --chown=dynamo: recipes/ /workspace/recipes/
+# Copy attribution files with correct ownership
+COPY --chown=dynamo: ATTRIBUTION* LICENSE /workspace/
 
 # Setup launch banner in common directory accessible to all users
 RUN --mount=type=bind,source=./container/launch_message.txt,target=/opt/dynamo/launch_message.txt \
diff --git a/container/README.md b/container/README.md
index f76ec3dd41..410d61367d 100644
--- a/container/README.md
+++ b/container/README.md
@@ -15,6 +15,8 @@ The NVIDIA Dynamo project uses containerized development and deployment to maint
   - `Dockerfile.trtllm` - For TensorRT-LLM inference backend
   - `Dockerfile.sglang` - For SGLang inference backend
   - `Dockerfile` - Base/standalone configuration
+  - `Dockerfile.frontend` - For Kubernetes Gateway API Inference Extension integration with EPP
+  - `Dockerfile.epp` - For building the Endpoint Picker (EPP) image
 
 ### Why Containerization?
 
@@ -192,6 +194,55 @@ The `build.sh --dev-image` option takes a dev image and then builds a local-dev
 ./build.sh --dev-image dynamo:latest-vllm --framework vllm --dry-run
 ```
 
+### Building the Frontend Image
+
+The frontend image is a specialized container that includes the Dynamo components (NATS, etcd, dynamo, NIXL, etc) along with the Endpoint Picker (EPP) for Kubernetes Gateway API Inference Extension integration. This image is primarily used for inference gateway deployments.
+
+**Step 1: Build the Custom Dynamo EPP Image**
+
+Follow the instructions in [`deploy/inference-gateway/README.md`](../deploy/inference-gateway/README.md) under "Build the custom EPP image" section. This process:
+- Clones the Gateway API Inference Extension repository
+- Applies Dynamo-specific patches for custom routing
+- Builds the Dynamo router as a static library
+- Creates a custom EPP image with integrated Dynamo routing capabilities
+
+**Step 2: Build the Dynamo Base Image**
+
+The base image contains the core Dynamo runtime components, NATS server, etcd, and Python dependencies:
+```bash
+# Build the base dev image (framework=none for frontend-only deployment)
+./build.sh --framework none --target dev
+```
+
+**Step 3: Build the Frontend Image**
+
+Now build the frontend image that combines the Dynamo base with the EPP:
+
+```bash
+# 2. Build the frontend image using the pre-built EPP
+docker buildx build --load --platform linux/amd64 \
+  --build-arg DYNAMO_BASE_IMAGE=dynamo:latest-none-dev \
+  --build-arg EPP_IMAGE={EPP_IMAGE_TAG} \
+  --build-arg PYTHON_VERSION=3.12 \
+  -f container/Dockerfile.frontend \
+  -t dynamo:latest-none-frontend \
+  .
+```
+#### Frontend Image Contents
+
+The frontend image includes:
+- **EPP (Endpoint Picker)**: Handles request routing and load balancing for inference gateway
+- **Dynamo Runtime**: Core platform components and routing logic
+- **NIXL**: NVIDIA InfiniBand Library for high-performance network communication
+- **Benchmarking Tools**: Performance testing utilities (aiperf, aiconfigurator, etc)
+- **Python Environment**: Virtual environment with all required dependencies
+- **NATS Server**: Message broker for Dynamo's distributed communication
+- **etcd**: Distributed key-value store for configuration and coordination
+
+#### Deployment
+
+The frontend image is designed for Kubernetes deployment with the Gateway API Inference Extension. See [`deploy/inference-gateway/README.md`](../deploy/inference-gateway/README.md) for complete deployment instructions using Helm charts.
+
 ### run.sh - Container Runtime Manager
 
 The `run.sh` script launches Docker containers with the appropriate configuration for development and inference workloads.

From e57f00f7ba579a64062abb8e15380256e6c9ba08 Mon Sep 17 00:00:00 2001
From: Tushar Sharma <tusharma@nvidia.com>
Date: Wed, 12 Nov 2025 10:00:15 -0800
Subject: [PATCH 12/14] fix trailing whitespace

Signed-off-by: Tushar Sharma <tusharma@nvidia.com>
---
 container/Dockerfile.frontend | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/container/Dockerfile.frontend b/container/Dockerfile.frontend
index cd538b5973..cc21dfb43a 100644
--- a/container/Dockerfile.frontend
+++ b/container/Dockerfile.frontend
@@ -72,7 +72,7 @@ RUN uv pip install \
     fi && \
     cd /opt/dynamo/benchmarks && \
     UV_GIT_LFS=1 uv pip install --no-cache .
-    
+
 # Install common and test dependencies
 RUN --mount=type=bind,source=./container/deps/requirements.txt,target=/tmp/requirements.txt \
     --mount=type=bind,source=./container/deps/requirements.test.txt,target=/tmp/requirements.test.txt \

From ddd1271159ff24f6e5e09834870e02f46597f73f Mon Sep 17 00:00:00 2001
From: Tushar Sharma <tusharma@nvidia.com>
Date: Wed, 12 Nov 2025 12:20:10 -0800
Subject: [PATCH 13/14] copy venv from dev stage, remove unnecessary packages

Signed-off-by: Tushar Sharma <tusharma@nvidia.com>
---
 container/Dockerfile.frontend | 54 ++++++++---------------------------
 1 file changed, 12 insertions(+), 42 deletions(-)

diff --git a/container/Dockerfile.frontend b/container/Dockerfile.frontend
index cc21dfb43a..5667cccb93 100644
--- a/container/Dockerfile.frontend
+++ b/container/Dockerfile.frontend
@@ -5,7 +5,6 @@
 ARG DYNAMO_BASE_IMAGE="dynamo:latest-none"
 ARG EPP_IMAGE="us-central1-docker.pkg.dev/k8s-staging-images/gateway-api-inference-extension/epp:v0.5.1-dirty"
 ARG PYTHON_VERSION=3.12
-ARG ENABLE_KVBM=false
 
 FROM ${DYNAMO_BASE_IMAGE} AS dynamo_base
 
@@ -13,14 +12,11 @@ FROM ${EPP_IMAGE} AS epp
 
 FROM nvcr.io/nvidia/base/ubuntu:noble-20250619 AS frontend
 
+ARG PYTHON_VERSION
 RUN apt-get update -y \
     && apt-get install -y --no-install-recommends \
-        # required for AIC perf files
+        # required for EPP
         ca-certificates \
-        git \
-        git-lfs \
-        # curl for downloading files
-        curl \
         libstdc++6 \
     && apt-get clean \
     && rm -rf /var/lib/apt/lists/*
@@ -49,39 +45,8 @@ COPY --from=dynamo_base /usr/bin/nats-server /usr/bin/nats-server
 COPY --from=dynamo_base /usr/local/bin/etcd/ /usr/local/bin/etcd/
 ENV PATH=/usr/local/bin/etcd/:$PATH
 
-# Install uv package manager
-COPY --from=ghcr.io/astral-sh/uv:latest /uv /uvx /bin/
-
-# Create and activate virtual environment
-ARG PYTHON_VERSION
-RUN uv venv /opt/dynamo/venv --python $PYTHON_VERSION
-
-ENV VIRTUAL_ENV=/opt/dynamo/venv \
-    PATH="/opt/dynamo/venv/bin:${PATH}"
-
-ARG ENABLE_KVBM
-# Copy built artifacts and workspace from dynamo_base (dev image)
-COPY --chown=dynamo: benchmarks/ /opt/dynamo/benchmarks/
-COPY --chown=dynamo: --from=dynamo_base /opt/dynamo/wheelhouse/ /opt/dynamo/wheelhouse/
-RUN uv pip install \
-    /opt/dynamo/wheelhouse/ai_dynamo_runtime*.whl \
-    /opt/dynamo/wheelhouse/ai_dynamo*any.whl \
-    /opt/dynamo/wheelhouse/nixl/nixl*.whl && \
-    if [ "$ENABLE_KVBM" = "true" ]; then \
-        uv pip install /opt/dynamo/wheelhouse/kvbm*.whl; \
-    fi && \
-    cd /opt/dynamo/benchmarks && \
-    UV_GIT_LFS=1 uv pip install --no-cache .
-
-# Install common and test dependencies
-RUN --mount=type=bind,source=./container/deps/requirements.txt,target=/tmp/requirements.txt \
-    --mount=type=bind,source=./container/deps/requirements.test.txt,target=/tmp/requirements.test.txt \
-    UV_GIT_LFS=1 uv pip install \
-        --no-cache \
-        --requirement /tmp/requirements.txt \
-        --requirement /tmp/requirements.test.txt
-
-# Copy tests, benchmarks, deploy and components for CI with correct ownership
+COPY --chown=dynamo: container/launch_message.txt /opt/dynamo/.launch_screen
+# Copy tests, benchmarks, deploy and components with correct ownership
 COPY --chown=dynamo: tests /workspace/tests
 COPY --chown=dynamo: examples /workspace/examples
 COPY --chown=dynamo: benchmarks /workspace/benchmarks
@@ -91,9 +56,14 @@ COPY --chown=dynamo: recipes/ /workspace/recipes/
 # Copy attribution files with correct ownership
 COPY --chown=dynamo: ATTRIBUTION* LICENSE /workspace/
 
-# Setup launch banner in common directory accessible to all users
-RUN --mount=type=bind,source=./container/launch_message.txt,target=/opt/dynamo/launch_message.txt \
-    sed '/^#\s/d' /opt/dynamo/launch_message.txt > /opt/dynamo/.launch_screen
+ENV VIRTUAL_ENV=/opt/dynamo/venv
+ENV PATH="/opt/dynamo/venv/bin:$PATH"
+# Copy virtual environment directly from dynamo_base (dev image)
+# This includes all installed packages: dynamo, nixl, requirements.txt, requirements.test.txt
+# Copy uv to system /bin
+COPY --from=dynamo_base /bin/uv /bin/uvx /bin/
+RUN uv python install $PYTHON_VERSION
+COPY --chown=dynamo: --from=dynamo_base /opt/dynamo/venv/ /opt/dynamo/venv/
 
 # Setup environment for all users
 USER root

From 9d43a4400b4078e199e5a1ee7cc60f7878df4326 Mon Sep 17 00:00:00 2001
From: Tushar Sharma <tusharma@nvidia.com>
Date: Wed, 12 Nov 2025 13:03:34 -0800
Subject: [PATCH 14/14] update entrypoint to use epp automatically

Signed-off-by: Tushar Sharma <tusharma@nvidia.com>
---
 container/Dockerfile.frontend | 6 +-----
 1 file changed, 1 insertion(+), 5 deletions(-)

diff --git a/container/Dockerfile.frontend b/container/Dockerfile.frontend
index 5667cccb93..ca65fdb301 100644
--- a/container/Dockerfile.frontend
+++ b/container/Dockerfile.frontend
@@ -40,11 +40,6 @@ ENV DYNAMO_HOME=/opt/dynamo
 WORKDIR /
 COPY --chown=dynamo: --from=epp /epp /epp
 
-# Copy NATS and ETCD from base stage
-COPY --from=dynamo_base /usr/bin/nats-server /usr/bin/nats-server
-COPY --from=dynamo_base /usr/local/bin/etcd/ /usr/local/bin/etcd/
-ENV PATH=/usr/local/bin/etcd/:$PATH
-
 COPY --chown=dynamo: container/launch_message.txt /opt/dynamo/.launch_screen
 # Copy tests, benchmarks, deploy and components with correct ownership
 COPY --chown=dynamo: tests /workspace/tests
@@ -73,5 +68,6 @@ RUN chmod 755 /opt/dynamo/.launch_screen && \
 
 USER dynamo
 
+ENTRYPOINT ["/epp"]
 CMD ["/bin/bash"]