Update MLflow package version

Dockerfile.sdk

@@ -29,10 +29,9 @@
 #
 
 # Base image on the minimum Triton container
-ARG BASE_IMAGE=nvcr.io/nvidia/tritonserver:25.08-py3-min
+ARG BASE_IMAGE=nvcr.io/nvidia/tritonserver:25.09-py3-min
 
 ARG TRITON_CLIENT_REPO_SUBDIR=clientrepo
-ARG TRITON_PA_REPO_SUBDIR=perfanalyzerrepo
 ARG TRITON_REPO_ORGANIZATION=http://github.com/triton-inference-server
 ARG TRITON_COMMON_REPO_TAG=main
 ARG TRITON_CORE_REPO_TAG=main
@@ -41,7 +40,6 @@ ARG TRITON_THIRD_PARTY_REPO_TAG=main
 ARG TRITON_ENABLE_GPU=ON
 ARG JAVA_BINDINGS_MAVEN_VERSION=3.8.4
 ARG JAVA_BINDINGS_JAVACPP_PRESETS_TAG=1.5.8
-ARG TRITON_PERF_ANALYZER_BUILD=1
 # DCGM version to install for Model Analyzer
 ARG DCGM_VERSION=4.4.0-1
 
@@ -97,7 +95,6 @@ RUN rm -f /usr/bin/python && \
 # Build the client library and examples
 ARG TRITON_REPO_ORGANIZATION
 ARG TRITON_CLIENT_REPO_SUBDIR
-ARG TRITON_PA_REPO_SUBDIR
 ARG TRITON_COMMON_REPO_TAG
 ARG TRITON_CORE_REPO_TAG
 ARG TRITON_CLIENT_REPO_TAG
@@ -106,14 +103,10 @@ ARG TRITON_ENABLE_GPU
 ARG JAVA_BINDINGS_MAVEN_VERSION
 ARG JAVA_BINDINGS_JAVACPP_PRESETS_TAG
 ARG TARGETPLATFORM
-ARG TRITON_PERF_ANALYZER_BUILD
-
-ENV TRITON_PERF_ANALYZER_BUILD=${TRITON_PERF_ANALYZER_BUILD}
 
 WORKDIR /workspace
 COPY TRITON_VERSION .
 COPY ${TRITON_CLIENT_REPO_SUBDIR} client
-COPY ${TRITON_PA_REPO_SUBDIR} perf_analyzer
 
 WORKDIR /workspace/client_build
 RUN cmake -DCMAKE_INSTALL_PREFIX=/workspace/install \
@@ -124,63 +117,11 @@ RUN cmake -DCMAKE_INSTALL_PREFIX=/workspace/install \
           -DTRITON_THIRD_PARTY_REPO_TAG=${TRITON_THIRD_PARTY_REPO_TAG} \
           -DTRITON_ENABLE_PERF_ANALYZER=OFF \
           -DTRITON_ENABLE_CC_HTTP=ON -DTRITON_ENABLE_CC_GRPC=ON \
-          -DTRITON_ENABLE_PYTHON_HTTP=OFF -DTRITON_ENABLE_PYTHON_GRPC=OFF \
+          -DTRITON_ENABLE_PYTHON_HTTP=ON -DTRITON_ENABLE_PYTHON_GRPC=ON \
           -DTRITON_ENABLE_JAVA_HTTP=ON \
           -DTRITON_ENABLE_EXAMPLES=ON -DTRITON_ENABLE_TESTS=ON \
           -DTRITON_ENABLE_GPU=${TRITON_ENABLE_GPU} /workspace/client
-RUN cmake --build . -v --parallel --target cc-clients java-clients
-
-# TODO: PA will rebuild the CC clients since it depends on it.
-# This should be optimized so that we do not have to build
-# the CC clients twice. Similarly, because the SDK expectation is
-# that PA is packaged with the python client, we hold off on building
-# the python client until now. Post-migration we should focus
-# effort on de-tangling these flows.
-WORKDIR /workspace/pa_build
-# NOTE: If TRITON_PERF_ANALYZER_BUILD=0, the Performance Analyzer (PA) binaries must already exist
-# in the path specified by the ARG TRITON_PA_REPO_SUBDIR.
-RUN if [ "$TRITON_PERF_ANALYZER_BUILD" = "1" ]; then \
-        cmake -DCMAKE_INSTALL_PREFIX=/workspace/install \
-          -DTRITON_VERSION=`cat /workspace/TRITON_VERSION` \
-          -DTRITON_REPO_ORGANIZATION=${TRITON_REPO_ORGANIZATION} \
-          -DTRITON_COMMON_REPO_TAG=${TRITON_COMMON_REPO_TAG} \
-          -DTRITON_CORE_REPO_TAG=${TRITON_CORE_REPO_TAG} \
-          -DTRITON_CLIENT_REPO_TAG=${TRITON_CLIENT_REPO_TAG} \
-          -DTRITON_THIRD_PARTY_REPO_TAG=${TRITON_THIRD_PARTY_REPO_TAG} \
-          -DTRITON_ENABLE_PERF_ANALYZER_C_API=ON \
-          -DTRITON_ENABLE_PERF_ANALYZER_TFS=ON \
-          -DTRITON_ENABLE_PERF_ANALYZER_TS=ON \
-          -DTRITON_ENABLE_PERF_ANALYZER_OPENAI=ON \
-          -DTRITON_ENABLE_CC_HTTP=ON \
-          -DTRITON_ENABLE_CC_GRPC=ON \
-          -DTRITON_ENABLE_PYTHON_HTTP=ON \
-          -DTRITON_ENABLE_PYTHON_GRPC=ON \
-          -DTRITON_PACKAGE_PERF_ANALYZER=ON \
-          -DTRITON_ENABLE_GPU=${TRITON_ENABLE_GPU} \
-        /workspace/perf_analyzer && \
-        cmake --build . -v --parallel --target perf-analyzer python-clients && \
-        pip3 install build && \
-        cd /workspace/perf_analyzer/genai-perf && \
-        python3 -m build --wheel --outdir /workspace/install/python; \
-    else \
-        ls /workspace/perf_analyzer/ && \
-        tar -xzf /workspace/perf_analyzer/perf_analyzer*.tar.gz -C /workspace/install/bin && \
-        echo "Perf Analyzer binaries was extracted and not build" && \
-        cmake -DCMAKE_INSTALL_PREFIX=/workspace/install \
-          -DTRITON_VERSION=`cat /workspace/TRITON_VERSION` \
-          -DTRITON_REPO_ORGANIZATION=${TRITON_REPO_ORGANIZATION} \
-          -DTRITON_COMMON_REPO_TAG=${TRITON_COMMON_REPO_TAG} \
-          -DTRITON_CLIENT_REPO_TAG=${TRITON_CLIENT_REPO_TAG} \
-          -DTRITON_THIRD_PARTY_REPO_TAG=${TRITON_THIRD_PARTY_REPO_TAG} \
-          -DTRITON_ENABLE_PYTHON_HTTP=ON \
-          -DTRITON_ENABLE_PYTHON_GRPC=ON \
-          -DTRITON_PACKAGE_PERF_ANALYZER=ON \
-          -DTRITON_ENABLE_GPU=${TRITON_ENABLE_GPU} \
-        /workspace/perf_analyzer && \
-        cmake --build . -v --parallel --target python-clients && \
-        mkdir -p /workspace/install/python && \
-        cp /workspace/perf_analyzer/genai_perf-*.whl /workspace/install/python/; \
-    fi
+RUN cmake --build . -v --parallel --target cc-clients java-clients python-clients
 
 # Install Java API Bindings
 RUN if [ "$TARGETPLATFORM" = "linux/amd64" ]; then \
@@ -226,7 +167,6 @@ RUN apt-get update && \
             python3-pip \
             python3-setuptools \
             python3-wheel \
-            software-properties-common \
             vim \
             wget && \
     pip3 install "grpcio<1.68" "grpcio-tools<1.68"
@@ -235,7 +175,6 @@ WORKDIR /workspace
 COPY TRITON_VERSION .
 COPY NVIDIA_Deep_Learning_Container_License.pdf .
 COPY --from=sdk_build /workspace/client/ client/
-COPY --from=sdk_build /workspace/perf_analyzer/ perf_analyzer/
 COPY --from=sdk_build /workspace/install/ install/
 RUN cd install && \
     export VERSION=`cat /workspace/TRITON_VERSION` && \
@@ -253,8 +192,6 @@ COPY --from=sdk_build /workspace/client/src/python/library/tests/* qa/python_cli
 # Install an image needed by the quickstart and other documentation.
 COPY qa/images/mug.jpg images/mug.jpg
 
-RUN pip3 install install/python/genai_perf-*.whl
-
 # Install the dependencies needed to run the client examples. These
 # are not needed for building but including them allows this image to
 # be used to run the client examples.
@@ -263,6 +200,9 @@ RUN pip3 install --upgrade "numpy<2" pillow attrdict && \
          "tritonclient-*linux*.whl" | xargs printf -- '%s[all]' | \
     xargs pip3 install --upgrade
 
+# Install GenAI-Perf
+RUN pip3 install genai-perf
+
 # Install DCGM
 RUN if [ "$TRITON_ENABLE_GPU" = "ON" ]; then \
         [ "$(uname -m)" != "x86_64" ] && arch="sbsa" || arch="x86_64" && \

README.md

@@ -27,11 +27,6 @@
 -->
 [![License](https://img.shields.io/badge/License-BSD3-lightgrey.svg)](https://opensource.org/licenses/BSD-3-Clause)
 
->[!WARNING]
->You are currently on the `main` branch which tracks under-development progress
->towards the next release. The current release is version [2.60.0](https://github.com/triton-inference-server/server/releases/latest)
->and corresponds to the 25.08 container release on NVIDIA GPU Cloud (NGC).
-
 # Triton Inference Server
 
 Triton Inference Server is an open source inference serving software that
@@ -61,7 +56,7 @@ Major features include:
 - Provides [Backend API](https://github.com/triton-inference-server/backend) that
   allows adding custom backends and pre/post processing operations
 - Supports writing custom backends in python, a.k.a.
-  [Python-based backends.](https://github.com/triton-inference-server/backend/blob/main/docs/python_based_backends.md#python-based-backends)
+  [Python-based backends.](https://github.com/triton-inference-server/backend/blob/r25.09/docs/python_based_backends.md#python-based-backends)
 - Model pipelines using
   [Ensembling](docs/user_guide/architecture.md#ensemble-models) or [Business
   Logic Scripting
@@ -90,16 +85,16 @@ Inference Server with the
 
 ```bash
 # Step 1: Create the example model repository
-git clone -b r25.08 https://github.com/triton-inference-server/server.git
+git clone -b r25.09 https://github.com/triton-inference-server/server.git
 cd server/docs/examples
 ./fetch_models.sh
 
 # Step 2: Launch triton from the NGC Triton container
-docker run --gpus=1 --rm --net=host -v ${PWD}/model_repository:/models nvcr.io/nvidia/tritonserver:25.08-py3 tritonserver --model-repository=/models --model-control-mode explicit --load-model densenet_onnx
+docker run --gpus=1 --rm --net=host -v ${PWD}/model_repository:/models nvcr.io/nvidia/tritonserver:25.09-py3 tritonserver --model-repository=/models --model-control-mode explicit --load-model densenet_onnx
 
 # Step 3: Sending an Inference Request
 # In a separate console, launch the image_client example from the NGC Triton SDK container
-docker run -it --rm --net=host nvcr.io/nvidia/tritonserver:25.08-py3-sdk /workspace/install/bin/image_client -m densenet_onnx -c 3 -s INCEPTION /workspace/images/mug.jpg
+docker run -it --rm --net=host nvcr.io/nvidia/tritonserver:25.09-py3-sdk /workspace/install/bin/image_client -m densenet_onnx -c 3 -s INCEPTION /workspace/images/mug.jpg
 
 # Inference should return the following
 Image '/workspace/images/mug.jpg':
@@ -172,10 +167,10 @@ configuration](docs/user_guide/model_configuration.md) for the model.
   [Python](https://github.com/triton-inference-server/python_backend), and more
 - Not all the above backends are supported on every platform supported by Triton.
   Look at the
-  [Backend-Platform Support Matrix](https://github.com/triton-inference-server/backend/blob/main/docs/backend_platform_support_matrix.md)
+  [Backend-Platform Support Matrix](https://github.com/triton-inference-server/backend/blob/r25.09/docs/backend_platform_support_matrix.md)
   to learn which backends are supported on your target platform.
 - Learn how to [optimize performance](docs/user_guide/optimization.md) using the
-  [Performance Analyzer](https://github.com/triton-inference-server/perf_analyzer/blob/main/README.md)
+  [Performance Analyzer](https://github.com/triton-inference-server/perf_analyzer/blob/r25.09/README.md)
   and
   [Model Analyzer](https://github.com/triton-inference-server/model_analyzer)
 - Learn how to [manage loading and unloading models](docs/user_guide/model_management.md) in
@@ -189,14 +184,14 @@ A Triton *client* application sends inference and other requests to Triton. The
 [Python and C++ client libraries](https://github.com/triton-inference-server/client)
 provide APIs to simplify this communication.
 
-- Review client examples for [C++](https://github.com/triton-inference-server/client/blob/main/src/c%2B%2B/examples),
-  [Python](https://github.com/triton-inference-server/client/blob/main/src/python/examples),
-  and [Java](https://github.com/triton-inference-server/client/blob/main/src/java/src/main/java/triton/client/examples)
+- Review client examples for [C++](https://github.com/triton-inference-server/client/blob/r25.09/src/c%2B%2B/examples),
+  [Python](https://github.com/triton-inference-server/client/blob/r25.09/src/python/examples),
+  and [Java](https://github.com/triton-inference-server/client/blob/r25.09/src/java/src/main/java/triton/client/examples)
 - Configure [HTTP](https://github.com/triton-inference-server/client#http-options)
   and [gRPC](https://github.com/triton-inference-server/client#grpc-options)
   client options
 - Send input data (e.g. a jpeg image) directly to Triton in the [body of an HTTP
-  request without any additional metadata](https://github.com/triton-inference-server/server/blob/main/docs/protocol/extension_binary_data.md#raw-binary-request)
+  request without any additional metadata](https://github.com/triton-inference-server/server/blob/r25.09/docs/protocol/extension_binary_data.md#raw-binary-request)
 
 ### Extend Triton
 
@@ -205,7 +200,7 @@ designed for modularity and flexibility
 
 - [Customize Triton Inference Server container](docs/customization_guide/compose.md) for your use case
 - [Create custom backends](https://github.com/triton-inference-server/backend)
-  in either [C/C++](https://github.com/triton-inference-server/backend/blob/main/README.md#triton-backend-api)
+  in either [C/C++](https://github.com/triton-inference-server/backend/blob/r25.09/README.md#triton-backend-api)
   or [Python](https://github.com/triton-inference-server/python_backend)
 - Create [decoupled backends and models](docs/user_guide/decoupled_models.md) that can send
   multiple responses for a request or not send any responses for a request

TRITON_VERSION

@@ -1 +1 @@
-2.61.0dev
+2.61.0

build.py

@@ -71,14 +71,14 @@
 #
 
 DEFAULT_TRITON_VERSION_MAP = {
-    "release_version": "2.61.0dev",
-    "triton_container_version": "25.09dev",
-    "upstream_container_version": "25.08",
+    "release_version": "2.61.0",
+    "triton_container_version": "25.09",
+    "upstream_container_version": "25.09",
     "ort_version": "1.23.0",
     "ort_openvino_version": "2025.3.0",
     "standalone_openvino_version": "2025.3.0",
     "dcgm_version": "4.4.0-1",
-    "vllm_version": "0.9.2",
+    "vllm_version": "0.10.1.1",
     "rhel_py_version": "3.12.3",
 }
 
@@ -1384,7 +1384,6 @@ def dockerfile_prepare_container_linux(argmap, backends, enable_gpu, target_mach
               libgoogle-perftools-dev \\
               libjemalloc-dev \\
               libnuma-dev \\
-              software-properties-common \\
               wget \\
               {backend_dependencies} \\
               python3-pip \\

deploy/aws/values.yaml

@@ -27,7 +27,7 @@
 replicaCount: 1
 
 image:
-  imageName: nvcr.io/nvidia/tritonserver:25.08-py3
+  imageName: nvcr.io/nvidia/tritonserver:25.09-py3
   pullPolicy: IfNotPresent
   modelRepositoryPath: s3://triton-inference-server-repository/model_repository
   numGpus: 1

deploy/fleetcommand/Chart.yaml

@@ -26,7 +26,7 @@
 
 apiVersion: v1
 # appVersion is the Triton version; update when changing release
-appVersion: 2.60.0"
+appVersion: 2.61.0"
 description: Triton Inference Server (Fleet Command)
 name: triton-inference-server
 # version is the Chart version; update when changing anything in the chart

deploy/fleetcommand/values.yaml

@@ -27,7 +27,7 @@
 replicaCount: 1
 
 image:
-  imageName: nvcr.io/nvidia/tritonserver:25.08-py3
+  imageName: nvcr.io/nvidia/tritonserver:25.09-py3
   pullPolicy: IfNotPresent
   numGpus: 1
   serverCommand: tritonserver
@@ -47,13 +47,13 @@ image:
     #
     # To set model control mode, uncomment and configure below
     # TODO: Fix the following url, it is invalid
-    # See https://github.com/triton-inference-server/server/blob/r25.08/docs/user_guide/model_management.md
+    # See https://github.com/triton-inference-server/server/blob/r25.09/docs/user_guide/model_management.md
     #  for more details
     #- --model-control-mode=explicit|poll|none
     #
     # Additional server args
     #
-    # see https://github.com/triton-inference-server/server/blob/r25.08/README.md
+    # see https://github.com/triton-inference-server/server/blob/r25.09/README.md
     #  for more details
 
 service:

deploy/gcp/values.yaml

@@ -27,7 +27,7 @@
 replicaCount: 1
 
 image:
-  imageName: nvcr.io/nvidia/tritonserver:25.08-py3
+  imageName: nvcr.io/nvidia/tritonserver:25.09-py3
   pullPolicy: IfNotPresent
   modelRepositoryPath: gs://triton-inference-server-repository/model_repository
   numGpus: 1

deploy/gke-marketplace-app/benchmark/perf-analyzer-script/triton_client.yaml

@@ -33,7 +33,7 @@ metadata:
   namespace: default
 spec:
   containers:
-  - image: nvcr.io/nvidia/tritonserver:25.08-py3-sdk
+  - image: nvcr.io/nvidia/tritonserver:25.09-py3-sdk
     imagePullPolicy: Always
     name: nv-triton-client
     securityContext:

deploy/gke-marketplace-app/server-deployer/build_and_push.sh

@@ -27,9 +27,9 @@
 
 export REGISTRY=gcr.io/$(gcloud config get-value project | tr ':' '/')
 export APP_NAME=tritonserver
-export MAJOR_VERSION=2.60
-export MINOR_VERSION=2.60.0
-export NGC_VERSION=25.08-py3
+export MAJOR_VERSION=2.61
+export MINOR_VERSION=2.61.0
+export NGC_VERSION=25.09-py3
 
 docker pull nvcr.io/nvidia/$APP_NAME:$NGC_VERSION
 

deploy/gke-marketplace-app/server-deployer/chart/triton/Chart.yaml

@@ -28,4 +28,4 @@ apiVersion: v1
 appVersion: "2.60"
 description: Triton Inference Server
 name: triton-inference-server
-version: 2.60.0
+version: 2.61.0

deploy/gke-marketplace-app/server-deployer/chart/triton/values.yaml

@@ -31,14 +31,14 @@ maxReplicaCount: 3
 tritonProtocol: HTTP
 # HPA GPU utilization autoscaling target
 HPATargetAverageValue: 85
-modelRepositoryPath: gs://triton_sample_models/25.08
-publishedVersion: '2.60.0'
+modelRepositoryPath: gs://triton_sample_models/25.09
+publishedVersion: '2.61.0'
 gcpMarketplace: true
 
 image:
   registry: gcr.io
   repository: nvidia-ngc-public/tritonserver
-  tag: 25.08-py3
+  tag: 25.09-py3
   pullPolicy: IfNotPresent
   # modify the model repository here to match your GCP storage bucket
   numGpus: 1

deploy/gke-marketplace-app/server-deployer/data-test/schema.yaml

@@ -27,7 +27,7 @@
 x-google-marketplace:
   schemaVersion: v2
   applicationApiVersion: v1beta1
-  publishedVersion: '2.60.0'
+  publishedVersion: '2.61.0'
   publishedVersionMetadata:
     releaseNote: >-
       Initial release.

deploy/gke-marketplace-app/server-deployer/schema.yaml

@@ -27,7 +27,7 @@
 x-google-marketplace:
   schemaVersion: v2
   applicationApiVersion: v1beta1
-  publishedVersion: '2.60.0'
+  publishedVersion: '2.61.0'
   publishedVersionMetadata:
     releaseNote: >-
       Initial release.
@@ -89,7 +89,7 @@ properties:
   modelRepositoryPath:
     type: string
     title: Bucket where models are stored. Please make sure the user/service account to create the GKE app has permission to this GCS bucket. Read Triton documentation on configs and formatting details, supporting TensorRT, TensorFlow, Pytorch, Onnx ... etc.
-    default: gs://triton_sample_models/25.08
+    default: gs://triton_sample_models/25.09
   image.ldPreloadPath:
     type: string
     title: Leave this empty by default. Triton allows users to create custom layers for backend such as TensorRT plugin, the compiled shared library must be provided via LD_PRELOAD environment variable.