chore: Templating Feedback Followup (#6125)

Signed-off-by: Dillon Cullinan <dcullinan@nvidia.com>

.devcontainer/README.md

@@ -143,7 +143,7 @@ Build the appropriate framework image (e.g., `dynamo:latest-vllm-local-dev`) fro
 ```bash
 # Single command approach (recommended)
 export FRAMEWORK=VLLM         # Note: any of VLLM, SGLANG, TRTLLM can be used
-python container/render.py --framework=${FRAMEWORK} --target=local-dev --short-output
+python container/render.py --framework=${FRAMEWORK} --target=local-dev --output-short-filename
 docker build --build-arg USER_UID=$(id -u) --build-arg USER_GID=$(id -g) -f container/rendered.Dockerfile .
 
 # Now you've created both dynamo:latest-vllm and dynamo:latest-vllm-local-dev
@@ -416,7 +416,7 @@ If you see errors like "container is not running" or "An error occurred setting
 
    # If missing, build the dev image first, then build local-dev
    export FRAMEWORK=VLLM  # Replace with VLLM, SGLANG, or TRTLLM
-   python container/render.py --framework=${FRAMEWORK} --target=local-dev --short-output
+   python container/render.py --framework=${FRAMEWORK} --target=local-dev --output-short-filename
    docker build --build-arg USER_UID=$(id -u) --build-arg USER_GID=$(id -g) -f container/rendered.Dockerfile .
    ```
 

.dockerignore

@@ -42,7 +42,9 @@
 **/target/*
 **/*safetensors
 container/Dockerfile*
-container/*.Dockerfile
+container/**/*.Dockerfile
+container/render.py
+container/context.yaml
 .venv
 .venv-docs
 

.github/actions/docker-build/action.yml

@@ -15,7 +15,7 @@ inputs:
     required: false
     default: 'linux/amd64'
   cuda_version:
-    description: 'Optional override for CUDA_VERSION build-arg'
+    description: 'CUDA version to use'
     required: true
   image_tag:
     description: 'Custom image tag (optional, defaults to framework:latest)'
@@ -77,7 +77,7 @@ runs:
             --platform=${{ inputs.platform }} \
             --cuda-version=${{ inputs.cuda_version }} \
             --show-result \
-            --short-output
+            --output-short-filename
         echo "::endgroup::"
     - name: Build EPP image
       if: ${{ inputs.target == 'frontend' }}

.github/actions/docker-remote-build/action.yml

@@ -14,7 +14,7 @@ inputs:
     required: false
     default: 'linux/amd64'
   cuda_version:
-    description: 'Optional override for CUDA_VERSION build-arg'
+    description: 'CUDA version to use'
     required: true
   image_tag:
     description: 'Custom image tag'
@@ -98,7 +98,7 @@ runs:
 
         # Create build logs directory
         mkdir -p build-logs
-        BUILD_LOG_FILE="build-logs/build-${{ inputs.framework }}-$(echo '${{ inputs.platform }}' | sed 's/linux\///').log"
+        BUILD_LOG_FILE="build-logs/build-${{ inputs.framework }}-${{ inputs.platform }}.log"
         echo "BUILD_LOG_FILE=${BUILD_LOG_FILE}" >> $GITHUB_ENV
         echo "📝 Build log will be saved to: ${BUILD_LOG_FILE}"
 
@@ -108,14 +108,14 @@ runs:
         CACHE_ARGS=""
         if [[ "${{ inputs.target }}" != "frontend" ]]; then
           CACHE_ARGS="--cache-to type=inline "
-          CACHE_ARGS+="--cache-from type=registry,ref=${ECR_HOSTNAME}/ai-dynamo/dynamo:${{ inputs.framework }}-cuda${CUDA_VERSION_MAJOR}-${PLATFORM##*/}-cache "
-          CACHE_ARGS+="--cache-from type=registry,ref=${ECR_HOSTNAME}/ai-dynamo/dynamo:main-${{ inputs.framework }}-cuda${CUDA_VERSION_MAJOR}-${PLATFORM##*/} "
-          CACHE_ARGS+="--cache-from type=registry,ref=${ECR_HOSTNAME}/ai-dynamo/dynamo:release-${{ inputs.framework }}-cuda${CUDA_VERSION_MAJOR}-${PLATFORM##*/}-cache "
+          CACHE_ARGS+="--cache-from type=registry,ref=${ECR_HOSTNAME}/ai-dynamo/dynamo:${{ inputs.framework }}-cuda${CUDA_VERSION_MAJOR}-${PLATFORM}-cache "
+          CACHE_ARGS+="--cache-from type=registry,ref=${ECR_HOSTNAME}/ai-dynamo/dynamo:main-${{ inputs.framework }}-cuda${CUDA_VERSION_MAJOR}-${PLATFORM} "
+          CACHE_ARGS+="--cache-from type=registry,ref=${ECR_HOSTNAME}/ai-dynamo/dynamo:release-${{ inputs.framework }}-cuda${CUDA_VERSION_MAJOR}-${PLATFORM}-cache "
           if [[ "$GITHUB_REF_NAME" =~ ^release ]]; then
             # Release branches also use release cache
-            CACHE_ARGS+="--cache-to type=registry,ref=${ECR_HOSTNAME}/ai-dynamo/dynamo:release-${{ inputs.framework }}-cuda${CUDA_VERSION_MAJOR}-${PLATFORM##*/}-cache,mode=max "
+            CACHE_ARGS+="--cache-to type=registry,ref=${ECR_HOSTNAME}/ai-dynamo/dynamo:release-${{ inputs.framework }}-cuda${CUDA_VERSION_MAJOR}-${PLATFORM}-cache,mode=max "
           elif [[ "$GITHUB_REF_NAME" == "main" ]]; then
-            CACHE_ARGS+="--cache-to type=registry,ref=${ECR_HOSTNAME}/ai-dynamo/dynamo:${{ inputs.framework }}-cuda${CUDA_VERSION_MAJOR}-${PLATFORM##*/}-cache,mode=max "
+            CACHE_ARGS+="--cache-to type=registry,ref=${ECR_HOSTNAME}/ai-dynamo/dynamo:${{ inputs.framework }}-cuda${CUDA_VERSION_MAJOR}-${PLATFORM}-cache,mode=max "
           fi
         fi
         echo "$CACHE_ARGS"

.github/filters.yaml

@@ -55,11 +55,7 @@ ignore:
   - '.github/release.yml'
   - '.github/copy-pr-bot.yaml'
   - '.github/dco.yml'
-  - 'container/Dockerfile.aws'
   - 'container/Dockerfile.docs'
-  - 'container/Dockerfile.epp'
-  - 'container/Dockerfile.frontend'
-  - 'container/Dockerfile.local_dev'
   - 'container/run.sh'
   - 'container/use-sccache.sh'
   - 'container/dev/**'
@@ -79,8 +75,6 @@ core:
   - '!**/*.rst'
   - '!**/*.txt'
   - *ci
-  - 'container/build.sh'
-  - 'container/Dockerfile'
   - 'container/render.py'
   - 'container/Dockerfile.template'
   - 'container/context.yaml'
@@ -124,7 +118,6 @@ planner:
 vllm:
   - '!**/*.md'
   - '!**/*.rst'
-  - 'container/Dockerfile.vllm'
   - 'container/deps/requirements.vllm.txt'
   - 'container/deps/vllm/**'
   - 'examples/backends/vllm/**'
@@ -134,7 +127,6 @@ vllm:
 sglang:
   - '!**/*.md'
   - '!**/*.rst'
-  - 'container/Dockerfile.sglang'
   - 'examples/backends/sglang/**'
   - 'components/src/dynamo/sglang/**'
   - 'container/templates/sglang_*'
@@ -142,7 +134,6 @@ sglang:
 trtllm:
   - '!**/*.md'
   - '!**/*.rst'
-  - 'container/Dockerfile.trtllm'
   - 'container/deps/trtllm/**'
   - 'examples/backends/trtllm/**'
   - 'components/src/dynamo/trtllm/**'
@@ -157,8 +148,6 @@ frontend:
   - 'lib/**'
   - '*.toml'
   - '*.lock'
-  - 'container/build.sh'
-  - 'container/Dockerfile'
   - 'container/deps/*'
   - 'components/src/dynamo/router/**'
   - 'components/src/dynamo/mocker/**'

.github/workflows/build-test-distribute-flavor.yml

@@ -139,13 +139,14 @@ jobs:
         shell: bash
         env:
           EXTRA_TAGS: ${{ inputs.extra_tags }}
-          PLATFORM: linux/${{ inputs.platform }}
+          CUDA_VERSION: ${{ inputs.cuda_version }}
         run: |
+          CUDA_VERSION_MAJOR=${CUDA_VERSION%%.*}
           if [ -n "$EXTRA_TAGS" ]; then
             RESULT=""
             while IFS= read -r tag; do
               if [ -n "$tag" ]; then
-                RESULT+="${{ secrets.AWS_ACCOUNT_ID }}.dkr.ecr.${{ secrets.AWS_DEFAULT_REGION }}.amazonaws.com/ai-dynamo/dynamo:${tag}-${{ inputs.platform }}"$'\n'
+                RESULT+="${{ secrets.AWS_ACCOUNT_ID }}.dkr.ecr.${{ secrets.AWS_DEFAULT_REGION }}.amazonaws.com/ai-dynamo/dynamo:${tag}-cuda${CUDA_VERSION_MAJOR}-${{ inputs.platform }}"$'\n'
               fi
             done <<< "$EXTRA_TAGS"
             echo "tags<<EOF" >> $GITHUB_OUTPUT
@@ -202,7 +203,7 @@ jobs:
               --platform=${{ inputs.platform }} \
               --cuda-version=${{ inputs.cuda_version }} \
               --show-result \
-              --short-output
+              --output-short-filename
       - name: Build Container
         id: build-image
         timeout-minutes: ${{ inputs.build_timeout_minutes }}

.github/workflows/container-validation-dynamo.yml

@@ -75,7 +75,7 @@ jobs:
               --framework=dynamo \
               --platform=amd64 \
               --show-result \
-              --short-output
+              --output-short-filename
       - name: Build image
         env:
           GITHUB_TOKEN: ${{ secrets.CI_TOKEN }}

container/Dockerfile.template

@@ -26,17 +26,32 @@
     {% endif %}
 {% endif %}
 
-# --- Framework Stages
+# --- SGLang Stages
 {% if framework == "sglang" %}
     # SGLang is the only framework without a `framework` target currently, needs special treatment
     {% include "templates/sglang_runtime.Dockerfile" %}
-{% elif target == "framework" and framework != "dynamo" %}
-    {% include "templates/" ~ framework ~ "_framework.Dockerfile" %}
-{% elif ( target == "runtime" or target == "dev" or target == "local-dev" ) and framework != "dynamo" %}
-    {% include "templates/" ~ framework ~ "_framework.Dockerfile" %}
-    {% include "templates/" ~ framework ~ "_runtime.Dockerfile" %}
 {% else %}
-# --- No framework stages included
+# --- No SGLANG stages included
+{% endif %}
+
+# --- VLLM Stages
+{% if framework == "vllm" %}
+    {% include "templates/vllm_framework.Dockerfile" %}
+    {% if target != "framework" %}
+        {% include "templates/vllm_runtime.Dockerfile" %}
+    {% endif %}
+{% else %}
+# --- No VLLM stages included
+{% endif %}
+
+# --- TRTLLM Stages
+{% if framework == "trtllm" %}
+    {% include "templates/trtllm_framework.Dockerfile" %}
+    {% if target != "framework" %}
+        {% include "templates/trtllm_runtime.Dockerfile" %}
+    {% endif %}
+{% else %}
+# --- No TRTLLM stages included
 {% endif %}
 
 {% if make_efa == true and target == "runtime" %}

container/README.md

@@ -4,18 +4,18 @@
 
 The NVIDIA Dynamo project uses containerized development and deployment to maintain consistent environments across different AI inference frameworks and deployment scenarios. This directory contains the tools for building and running Dynamo containers:
 
-### Core Components
-
-- **`render.py`** - A render script used to generate Dockerfiles for AI inference frameworks (vLLM, TensorRT-LLM, SGLang). The generated Dockerfile includes the needed multi-stage steps for development vs production configurations
-
-- **`run.sh`** - A container runtime manager that launches Docker containers with proper GPU access, volume mounts, and environment configurations. It supports different development workflows from root-based legacy setups to user-based development environments.
-
-## Rendering Requirements:
+### Rendering Requirements:
 - Python
 - Python Packages:
   - pyyaml
   - jinja2
 
+### Core Components
+
+- **`render.py`** - A render script used to generate Dockerfiles for AI inference frameworks (vLLM, TensorRT-LLM, SGLang) and the frontend image. The generated Dockerfile includes the needed multi-stage steps for development vs production configurations.
+
+- **`run.sh`** - A container runtime manager that launches Docker containers with proper GPU access, volume mounts, and environment configurations. It supports different development workflows from root-based legacy setups to user-based development environments.
+
 ### Stage Summary for Frameworks
 
 <details>
@@ -115,7 +115,7 @@ The `run.sh` script and rendering scripts are convenience that simplify common D
 ### 1. runtime target (runs as non-root dynamo user):
 ```bash
 # Build runtime image
-python container/render.py --framework vllm --target runtime --short-output
+python container/render.py --framework vllm --target runtime --output-short-filename
 docker build -t dynamo:latest-vllm-runtime -f rendered.Dockerfile .
 
 # Run runtime container
@@ -226,16 +226,16 @@ Note: `uv` commands set `UV_CACHE_DIR` per `RUN` so `uv` always uses the same pa
 
 ```bash
 # Build vLLM dev image called dynamo:latest-vllm (default). This runs as root and is for development.
-python container/render.py --framework=vllm --target=dev --short-output
+python container/render.py --framework=vllm --target=dev --output-short-filename
 docker build -t dynamo:latest-vllm-dev -f rendered.Dockerfile .
 
 # Build a local-dev image. The local-dev image will run as `dynamo` with UID/GID matched to your host user,
 # which is useful when mounting partitions for development.
-python container/render.py --framework=vllm --target=local-dev --short-output
+python container/render.py --framework=vllm --target=local-dev --output-short-filename
 docker build --build-arg USER_UID=$(id -u) --build-arg USER_GID=$(id -g) -f container/rendered.Dockerfile -t dynamo:latest-vllm-local-dev .
 
 # Build TensorRT-LLM development image called dynamo:latest-trtllm
-python container/render.py --framework=trtllm --target=runtime --short-output
+python container/render.py --framework=trtllm --target=runtime --output-short-filename
 docker build -t dynamo:latest-trtllm-runtime -f rendered.Dockerfile .
 ```
 
@@ -261,7 +261,7 @@ EPP_IMAGE="dynamo/dynamo-epp:${EPP_GIT_TAG}"
 **Build Frontend Image**
 ```bash
 # Build the frontend image (automatically builds EPP image as a dependency)
-python container/render.py --framework=dynamo --target=frontend --short-output
+python container/render.py --framework=dynamo --target=frontend --output-short-filename
 docker build -t dynamo:frontend --build-arg EPP_IMAGE=${EPP_IMAGE} -f rendered.Dockerfile .
 ```
 
@@ -421,7 +421,7 @@ See Docker documentation for custom network creation and management.
 ### Development Workflow
 ```bash
 # 1. Build local-dev image (builds runtime, then dev as intermediate, then local-dev as final image)
-python container/render.py --framework=vllm --target=local-dev --short-output
+python container/render.py --framework=vllm --target=local-dev --output-short-filename
 docker build --build-arg USER_UID=$(id -u) --build-arg USER_GID=$(id -g) -f container/rendered.Dockerfile -t dynamo:latest-vllm-local-dev .
 
 # 2. Run development container using the local-dev image
@@ -439,7 +439,7 @@ python -m dynamo.vllm --model Qwen/Qwen3-0.6B --gpu-memory-utilization 0.20 &
 ### Production Workflow
 ```bash
 # 1. Build production runtime image (runs as non-root dynamo user)
-python container/render.py --framework=vllm --target=runtime --short-output
+python container/render.py --framework=vllm --target=runtime --output-short-filename
 docker build -t dynamo:latest-vllm-runtime -f rendered.Dockerfile .
 
 # 2. Run production container as non-root dynamo user
@@ -449,7 +449,7 @@ container/run.sh --image dynamo:latest-vllm-runtime --gpus all -v $HOME/.cache:/
 ### Testing Workflow
 ```bash
 # 1. Build dev image
-python container/render.py --framework=vllm --target=dev --short-output
+python container/render.py --framework=vllm --target=dev --output-short-filename
 docker build -t dynamo:latest-vllm-dev -f rendered.Dockerfile .
 
 # 2. Run tests with network isolation for reproducible results (no -it needed for CI)

container/context.yaml

@@ -34,6 +34,10 @@ vllm:
   base_image: nvcr.io/nvidia/cuda-dl-base
   runtime_image: nvcr.io/nvidia/cuda
   runtime_image_tag: ${CUDA_VERSION}.0-runtime-ubuntu24.04
+  cuda12.9:
+    base_image_tag: 25.06-cuda12.9-devel-ubuntu24.04
+  cuda13.0:
+    base_image_tag: 25.11-cuda13.0-devel-ubuntu24.04
   vllm_ref: v0.14.1
   flashinf_ref: v0.5.3
   lmcache_ref: 0.3.12
@@ -41,10 +45,6 @@ vllm:
   enable_media_ffmpeg: "true"
   enable_gpu_memory_service: "true"
   enable_kvbm: "true"
-  cuda12.9:
-    base_image_tag: 25.06-cuda12.9-devel-ubuntu24.04
-  cuda13.0:
-    base_image_tag: 25.11-cuda13.0-devel-ubuntu24.04
 
 sglang:
   base_image: nvcr.io/nvidia/cuda-dl-base

container/render.py

@@ -4,7 +4,6 @@
 
 import argparse
 import re
-import sys
 from pathlib import Path
 
 import yaml
@@ -41,9 +40,9 @@ def parse_args():
     )
     parser.add_argument("--make-efa", action="store_true", help="Enable AWS EFA")
     parser.add_argument(
-        "--short-output",
+        "--output-short-filename",
         action="store_true",
-        help="Output filename is just rendered.Dockerfile",
+        help="Output filename is rendered.Dockerfile instead of <framework>-<target>-cuda<cuda_version>-<arch>-rendered.Dockerfile",
     )
     parser.add_argument(
         "--show-result",
@@ -55,7 +54,23 @@ def parse_args():
 
 
 def validate_args(args):
-    # TODO: Add validation logic
+    valid_inputs = {
+        "vllm": {"runtime", "dev", "local-dev", "framework", "wheel_builder", "base"},
+        "trtllm": {"runtime", "dev", "local-dev", "framework", "wheel_builder", "base"},
+        "sglang": {"runtime", "dev", "local-dev", "wheel_builder", "base"},
+        "dynamo": {"runtime", "dev", "local-dev", "frontend", "wheel_builder", "base"},
+    }
+
+    if args.framework in valid_inputs:
+        if args.target in valid_inputs[args.framework]:
+            return
+        raise ValueError(
+            f"Invalid input combination: [framework={args.framework},target={args.target}]"
+        )
+
+    raise ValueError(
+        f"Invalid input combination: [framework={args.framework},target={args.target}]"
+    )
     return
 
 
@@ -75,7 +90,7 @@ def render(args, context, script_dir):
     # Replace all instances of 3+ newlines with 2 newlines
     cleaned = re.sub(r"\n{3,}", "\n\n", rendered)
 
-    if args.short_output:
+    if args.output_short_filename:
         filename = "rendered.Dockerfile"
     else:
         filename = f"{args.framework}-{args.target}-cuda{args.cuda_version}-{args.platform}-rendered.Dockerfile"
@@ -98,7 +113,7 @@ def render(args, context, script_dir):
 def main():
     args = parse_args()
     validate_args(args)
-    script_dir = Path(sys.argv[0]).parent
+    script_dir = Path(__file__).parent
     with open(f"{script_dir}/context.yaml", "r") as f:
         context = yaml.safe_load(f)
 

container/templates/args.Dockerfile

@@ -2,6 +2,7 @@
 # SPDX-FileCopyrightText: Copyright (c) 2024-2026 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
 # SPDX-License-Identifier: Apache-2.0
 #}
+# === BEGIN templates/args.Dockerfile ===
 ##########################
 #### Build Arguments #####
 ##########################

container/templates/aws.Dockerfile

@@ -2,6 +2,7 @@
 # SPDX-FileCopyrightText: Copyright (c) 2024-2026 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
 # SPDX-License-Identifier: Apache-2.0
 #}
+# === BEGIN templates/aws.Dockerfile ===
 #############################
 ########## AWS EFA ##########
 #############################

container/templates/dev.Dockerfile

@@ -2,6 +2,7 @@
 # SPDX-FileCopyrightText: Copyright (c) 2024-2026 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
 # SPDX-License-Identifier: Apache-2.0
 #}
+# === BEGIN templates/dev.Dockerfile ===
 # ======================================================================
 # STAGE: dynamo_tools for developers
 # ======================================================================
@@ -126,9 +127,9 @@ RUN --mount=type=cache,target=/var/cache/apt,sharing=locked \
 # Add NVIDIA devtools repository and install development tools (nsight-systems).
 # Cache apt downloads; sharing=locked avoids apt/dpkg races with concurrent builds.
 RUN --mount=type=cache,target=/var/cache/apt,sharing=locked \
-    wget -qO - "https://developer.download.nvidia.com/devtools/repos/ubuntu2404/amd64/nvidia.pub" \
+    wget -qO - "https://developer.download.nvidia.com/devtools/repos/ubuntu2404/${ARCH}/nvidia.pub" \
         | gpg --dearmor -o /etc/apt/keyrings/nvidia-devtools.gpg && \
-    echo "deb [signed-by=/etc/apt/keyrings/nvidia-devtools.gpg] https://developer.download.nvidia.com/devtools/repos/ubuntu2404/amd64 /" \
+    echo "deb [signed-by=/etc/apt/keyrings/nvidia-devtools.gpg] https://developer.download.nvidia.com/devtools/repos/ubuntu2404/${ARCH} /" \
         | tee /etc/apt/sources.list.d/nvidia-devtools.list && \
     apt-get update && \
     apt-get install -y --no-install-recommends nsight-systems-2025.5.1 && \

container/templates/dynamo_base.Dockerfile

@@ -2,6 +2,7 @@
 # SPDX-FileCopyrightText: Copyright (c) 2024-2026 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
 # SPDX-License-Identifier: Apache-2.0
 #}
+# === BEGIN templates/dynamo_base.Dockerfile ===
 ##################################
 ########## Base Image ############
 ##################################

container/templates/dynamo_runtime.Dockerfile

@@ -2,6 +2,7 @@
 # SPDX-FileCopyrightText: Copyright (c) 2024-2026 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
 # SPDX-License-Identifier: Apache-2.0
 #}
+# === BEGIN templates/dynamo_runtime.Dockerfile ===
 #######################################
 ########## Runtime image ##############
 #######################################

container/templates/frontend.Dockerfile

@@ -2,6 +2,7 @@
 # SPDX-FileCopyrightText: Copyright (c) 2024-2026 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
 # SPDX-License-Identifier: Apache-2.0
 #}
+# === BEGIN templates/frontend.Dockerfile ===
 ##############################################
 ########## Frontend entrypoint image #########
 ##############################################

container/templates/local_dev.Dockerfile

@@ -2,6 +2,7 @@
 # SPDX-FileCopyrightText: Copyright (c) 2024-2026 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
 # SPDX-License-Identifier: Apache-2.0
 #}
+# === BEGIN templates/local_dev.Dockerfile ===
 # ======================================================================
 # TARGET: local-dev (non-root development with UID/GID remapping)
 # ======================================================================

container/templates/sglang_runtime.Dockerfile

@@ -2,6 +2,7 @@
 # SPDX-FileCopyrightText: Copyright (c) 2024-2026 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
 # SPDX-License-Identifier: Apache-2.0
 #}
+# === BEGIN templates/sglang_runtime.Dockerfile ===
 ##################################
 ########## Runtime Image #########
 ##################################

container/templates/trtllm_framework.Dockerfile

@@ -2,6 +2,8 @@
 # SPDX-FileCopyrightText: Copyright (c) 2024-2026 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
 # SPDX-License-Identifier: Apache-2.0
 #}
+# === BEGIN templates/trtllm_framework.Dockerfile ===
+
 # Copy artifacts from NGC PyTorch image
 FROM ${BASE_IMAGE}:${BASE_IMAGE_TAG} AS pytorch_base