From 44135039057b5d856c1b2ff430d6d3de777bb73d Mon Sep 17 00:00:00 2001 From: Misha Chornyi Date: Thu, 14 Aug 2025 19:52:57 -0700 Subject: [PATCH 1/2] versions: Update versions for 25.08 release --- Dockerfile.sdk | 2 +- README.md | 6 +++--- TRITON_VERSION | 2 +- build.py | 8 ++++---- deploy/aws/values.yaml | 2 +- deploy/fleetcommand/Chart.yaml | 2 +- deploy/fleetcommand/values.yaml | 6 +++--- deploy/gcp/values.yaml | 2 +- .../benchmark/perf-analyzer-script/triton_client.yaml | 2 +- .../gke-marketplace-app/server-deployer/build_and_push.sh | 6 +++--- .../server-deployer/chart/triton/Chart.yaml | 4 ++-- .../server-deployer/chart/triton/values.yaml | 6 +++--- .../server-deployer/data-test/schema.yaml | 2 +- deploy/gke-marketplace-app/server-deployer/schema.yaml | 6 +++--- deploy/gke-marketplace-app/trt-engine/README.md | 6 +++--- deploy/k8s-onprem/values.yaml | 2 +- deploy/oci/values.yaml | 2 +- python/openai/README.md | 2 +- qa/common/gen_jetson_trt_models | 2 +- qa/common/gen_qa_custom_ops | 2 +- qa/common/gen_qa_model_repository | 2 +- 21 files changed, 37 insertions(+), 37 deletions(-) diff --git a/Dockerfile.sdk b/Dockerfile.sdk index f7bbf64432..3dc07722a5 100644 --- a/Dockerfile.sdk +++ b/Dockerfile.sdk @@ -29,7 +29,7 @@ # # Base image on the minimum Triton container -ARG BASE_IMAGE=nvcr.io/nvidia/tritonserver:25.07-py3-min +ARG BASE_IMAGE=nvcr.io/nvidia/tritonserver:25.08-py3-min ARG TRITON_CLIENT_REPO_SUBDIR=clientrepo ARG TRITON_PA_REPO_SUBDIR=perfanalyzerrepo diff --git a/README.md b/README.md index eb610cc0dd..731557cb2b 100644 --- a/README.md +++ b/README.md @@ -90,16 +90,16 @@ Inference Server with the ```bash # Step 1: Create the example model repository -git clone -b r25.07 https://github.com/triton-inference-server/server.git +git clone -b r25.08 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:25.07-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:25.08-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:25.07-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:25.08-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': diff --git a/TRITON_VERSION b/TRITON_VERSION index 7b28f253ee..54bbd78ddc 100644 --- a/TRITON_VERSION +++ b/TRITON_VERSION @@ -1 +1 @@ -2.60.0dev +2.60.0 diff --git a/build.py b/build.py index 9cebec7a06..5ab53168da 100755 --- a/build.py +++ b/build.py @@ -71,13 +71,13 @@ # DEFAULT_TRITON_VERSION_MAP = { - "release_version": "2.60.0dev", - "triton_container_version": "25.08dev", - "upstream_container_version": "25.07", + "release_version": "2.60.0", + "triton_container_version": "25.08", + "upstream_container_version": "25.08", "ort_version": "1.23.0", "ort_openvino_version": "2025.2.0", "standalone_openvino_version": "2025.2.0", - "dcgm_version": "4.2.3-2", + "dcgm_version": "4.4.0-1", "vllm_version": "0.9.2", "rhel_py_version": "3.12.3", } diff --git a/deploy/aws/values.yaml b/deploy/aws/values.yaml index eb9afb78e0..6ec8264ae1 100644 --- a/deploy/aws/values.yaml +++ b/deploy/aws/values.yaml @@ -27,7 +27,7 @@ replicaCount: 1 image: - imageName: nvcr.io/nvidia/tritonserver:25.07-py3 + imageName: nvcr.io/nvidia/tritonserver:25.08-py3 pullPolicy: IfNotPresent modelRepositoryPath: s3://triton-inference-server-repository/model_repository numGpus: 1 diff --git a/deploy/fleetcommand/Chart.yaml b/deploy/fleetcommand/Chart.yaml index 1b1af554ee..5aebbe4957 100644 --- a/deploy/fleetcommand/Chart.yaml +++ b/deploy/fleetcommand/Chart.yaml @@ -26,7 +26,7 @@ apiVersion: v1 # appVersion is the Triton version; update when changing release -appVersion: "2.59.1" +appVersion: 2.60.0" description: Triton Inference Server (Fleet Command) name: triton-inference-server # version is the Chart version; update when changing anything in the chart diff --git a/deploy/fleetcommand/values.yaml b/deploy/fleetcommand/values.yaml index 03569ab343..94aa860e77 100644 --- a/deploy/fleetcommand/values.yaml +++ b/deploy/fleetcommand/values.yaml @@ -27,7 +27,7 @@ replicaCount: 1 image: - imageName: nvcr.io/nvidia/tritonserver:25.07-py3 + imageName: nvcr.io/nvidia/tritonserver:25.08-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/r25.07/docs/user_guide/model_management.md + # See https://github.com/triton-inference-server/server/blob/r25.08/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/r25.07/README.md + # see https://github.com/triton-inference-server/server/blob/r25.08/README.md # for more details service: diff --git a/deploy/gcp/values.yaml b/deploy/gcp/values.yaml index 55d0919205..027cf85819 100644 --- a/deploy/gcp/values.yaml +++ b/deploy/gcp/values.yaml @@ -27,7 +27,7 @@ replicaCount: 1 image: - imageName: nvcr.io/nvidia/tritonserver:25.07-py3 + imageName: nvcr.io/nvidia/tritonserver:25.08-py3 pullPolicy: IfNotPresent modelRepositoryPath: gs://triton-inference-server-repository/model_repository numGpus: 1 diff --git a/deploy/gke-marketplace-app/benchmark/perf-analyzer-script/triton_client.yaml b/deploy/gke-marketplace-app/benchmark/perf-analyzer-script/triton_client.yaml index 951f04898e..842362d6ce 100644 --- a/deploy/gke-marketplace-app/benchmark/perf-analyzer-script/triton_client.yaml +++ b/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:25.07-py3-sdk + - image: nvcr.io/nvidia/tritonserver:25.08-py3-sdk imagePullPolicy: Always name: nv-triton-client securityContext: diff --git a/deploy/gke-marketplace-app/server-deployer/build_and_push.sh b/deploy/gke-marketplace-app/server-deployer/build_and_push.sh index 659e3a40eb..8ff6eeb631 100755 --- a/deploy/gke-marketplace-app/server-deployer/build_and_push.sh +++ b/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=2.59 -export MINOR_VERSION=2.59.1 -export NGC_VERSION=25.07-py3 +export MAJOR_VERSION=2.60 +export MINOR_VERSION=2.60.0 +export NGC_VERSION=25.08-py3 docker pull nvcr.io/nvidia/$APP_NAME:$NGC_VERSION diff --git a/deploy/gke-marketplace-app/server-deployer/chart/triton/Chart.yaml b/deploy/gke-marketplace-app/server-deployer/chart/triton/Chart.yaml index 218a45c7a5..04d80cbfd4 100644 --- a/deploy/gke-marketplace-app/server-deployer/chart/triton/Chart.yaml +++ b/deploy/gke-marketplace-app/server-deployer/chart/triton/Chart.yaml @@ -25,7 +25,7 @@ # OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. apiVersion: v1 -appVersion: "2.59" +appVersion: "2.60" description: Triton Inference Server name: triton-inference-server -version: 2.59.1 +version: 2.60.0 diff --git a/deploy/gke-marketplace-app/server-deployer/chart/triton/values.yaml b/deploy/gke-marketplace-app/server-deployer/chart/triton/values.yaml index 94e5132d4d..7c7862d3f7 100644 --- a/deploy/gke-marketplace-app/server-deployer/chart/triton/values.yaml +++ b/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/25.07 -publishedVersion: '2.59.1' +modelRepositoryPath: gs://triton_sample_models/25.08 +publishedVersion: '2.60.0' gcpMarketplace: true image: registry: gcr.io repository: nvidia-ngc-public/tritonserver - tag: 25.07-py3 + tag: 25.08-py3 pullPolicy: IfNotPresent # modify the model repository here to match your GCP storage bucket numGpus: 1 diff --git a/deploy/gke-marketplace-app/server-deployer/data-test/schema.yaml b/deploy/gke-marketplace-app/server-deployer/data-test/schema.yaml index 949612450b..7a1774155f 100644 --- a/deploy/gke-marketplace-app/server-deployer/data-test/schema.yaml +++ b/deploy/gke-marketplace-app/server-deployer/data-test/schema.yaml @@ -27,7 +27,7 @@ x-google-marketplace: schemaVersion: v2 applicationApiVersion: v1beta1 - publishedVersion: '2.59.1' + publishedVersion: '2.60.0' publishedVersionMetadata: releaseNote: >- Initial release. diff --git a/deploy/gke-marketplace-app/server-deployer/schema.yaml b/deploy/gke-marketplace-app/server-deployer/schema.yaml index 5a2efd9e6b..607aafe4a5 100644 --- a/deploy/gke-marketplace-app/server-deployer/schema.yaml +++ b/deploy/gke-marketplace-app/server-deployer/schema.yaml @@ -27,7 +27,7 @@ x-google-marketplace: schemaVersion: v2 applicationApiVersion: v1beta1 - publishedVersion: '2.59.1' + publishedVersion: '2.60.0' publishedVersionMetadata: releaseNote: >- Initial release. @@ -88,8 +88,8 @@ properties: default: 85 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, Pytorch, Onnx ... etc. - default: gs://triton_sample_models/25.07 + 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 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. diff --git a/deploy/gke-marketplace-app/trt-engine/README.md b/deploy/gke-marketplace-app/trt-engine/README.md index c80da18f89..4651799daf 100644 --- a/deploy/gke-marketplace-app/trt-engine/README.md +++ b/deploy/gke-marketplace-app/trt-engine/README.md @@ -33,7 +33,7 @@ ``` docker run --gpus all -it --network host \ --shm-size=1g --ulimit memlock=-1 --ulimit stack=67108864 \ - -v ~:/scripts nvcr.io/nvidia/tensorrt:25.07-py3 + -v ~:/scripts nvcr.io/nvidia/tensorrt:25.08-py3 pip install onnx six torch tf2onnx tensorflow @@ -57,7 +57,7 @@ mkdir -p engines python3 builder.py -m models/fine-tuned/bert_tf_ckpt_large_qa_squad2_amp_128_v19.03.1/model.ckpt -o engines/bert_large_int8_bs1_s128.engine -b 1 -s 128 -c models/fine-tuned/bert_tf_ckpt_large_qa_squad2_amp_128_v19.03.1/ -v models/fine-tuned/bert_tf_ckpt_large_qa_squad2_amp_128_v19.03.1/vocab.txt --int8 --fp16 --strict --calib-num 1 -iln -imh -gsutil cp bert_large_int8_bs1_s128.engine gs://triton_sample_models/25.07/bert/1/model.plan +gsutil cp bert_large_int8_bs1_s128.engine gs://triton_sample_models/25.08/bert/1/model.plan ``` -For each Triton upgrade, container version used to generate the model, and the model path in GCS `gs://triton_sample_models/25.07/` should be updated accordingly with the correct version. +For each Triton upgrade, container version used to generate the model, and the model path in GCS `gs://triton_sample_models/25.08/` should be updated accordingly with the correct version. diff --git a/deploy/k8s-onprem/values.yaml b/deploy/k8s-onprem/values.yaml index 84d6c62f28..402a52635f 100644 --- a/deploy/k8s-onprem/values.yaml +++ b/deploy/k8s-onprem/values.yaml @@ -30,7 +30,7 @@ tags: openshift: false image: - imageName: nvcr.io/nvidia/tritonserver:25.07-py3 + imageName: nvcr.io/nvidia/tritonserver:25.08-py3 pullPolicy: IfNotPresent modelRepositoryServer: < Replace with the IP Address of your file server > modelRepositoryPath: /srv/models diff --git a/deploy/oci/values.yaml b/deploy/oci/values.yaml index 093791e1be..7261b9c0a3 100644 --- a/deploy/oci/values.yaml +++ b/deploy/oci/values.yaml @@ -27,7 +27,7 @@ replicaCount: 1 image: - imageName: nvcr.io/nvidia/tritonserver:25.07-py3 + imageName: nvcr.io/nvidia/tritonserver:25.08-py3 pullPolicy: IfNotPresent modelRepositoryPath: s3://https://.compat.objectstorage..oraclecloud.com:443/triton-inference-server-repository numGpus: 1 diff --git a/python/openai/README.md b/python/openai/README.md index 3feebc0980..915dbd923b 100644 --- a/python/openai/README.md +++ b/python/openai/README.md @@ -58,7 +58,7 @@ docker run -it --net=host --gpus all --rm \ -v ${HOME}/.cache/huggingface:/root/.cache/huggingface \ -e HF_TOKEN \ - nvcr.io/nvidia/tritonserver:25.07-vllm-python-py3 + nvcr.io/nvidia/tritonserver:25.08-vllm-python-py3 ``` 2. Launch the OpenAI-compatible Triton Inference Server: diff --git a/qa/common/gen_jetson_trt_models b/qa/common/gen_jetson_trt_models index 956ba03955..bf3281bc9b 100755 --- a/qa/common/gen_jetson_trt_models +++ b/qa/common/gen_jetson_trt_models @@ -34,7 +34,7 @@ # Make all generated files accessible outside of container umask 0000 # Set the version of the models -TRITON_VERSION=${TRITON_VERSION:=25.07} +TRITON_VERSION=${TRITON_VERSION:=25.08} # Set the CUDA device to use CUDA_DEVICE=${RUNNER_ID:=0} # Set TensorRT image diff --git a/qa/common/gen_qa_custom_ops b/qa/common/gen_qa_custom_ops index 6b3f349f44..a716c0e517 100755 --- a/qa/common/gen_qa_custom_ops +++ b/qa/common/gen_qa_custom_ops @@ -37,7 +37,7 @@ ## ############################################################################ -TRITON_VERSION=${TRITON_VERSION:=25.07} +TRITON_VERSION=${TRITON_VERSION:=25.08} NVIDIA_UPSTREAM_VERSION=${NVIDIA_UPSTREAM_VERSION:=$TRITON_VERSION} PYTORCH_IMAGE=${PYTORCH_IMAGE:=nvcr.io/nvidia/pytorch:$NVIDIA_UPSTREAM_VERSION-py3} UBUNTU_IMAGE=${UBUNTU_IMAGE:=ubuntu:24.04} diff --git a/qa/common/gen_qa_model_repository b/qa/common/gen_qa_model_repository index b70af4b6fc..67e2adb902 100755 --- a/qa/common/gen_qa_model_repository +++ b/qa/common/gen_qa_model_repository @@ -48,7 +48,7 @@ ## ############################################################################ -TRITON_VERSION=${TRITON_VERSION:=25.07} +TRITON_VERSION=${TRITON_VERSION:=25.08} # ONNX. Use ONNX_OPSET 0 to use the default for ONNX version ONNX_VERSION=1.16.1 From b432698b891df6b0e72305efa9661adde8c80a3e Mon Sep 17 00:00:00 2001 From: Misha Chornyi Date: Thu, 14 Aug 2025 20:40:39 -0700 Subject: [PATCH 2/2] Update DCGM version values --- Dockerfile.sdk | 2 +- compose.py | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) diff --git a/Dockerfile.sdk b/Dockerfile.sdk index 3dc07722a5..c517be2918 100644 --- a/Dockerfile.sdk +++ b/Dockerfile.sdk @@ -43,7 +43,7 @@ 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=4.2.3-2 +ARG DCGM_VERSION=4.4.0-1 ARG NVIDIA_TRITON_SERVER_SDK_VERSION=unknown ARG NVIDIA_BUILD_ID=unknown diff --git a/compose.py b/compose.py index ad177d56b2..e2321cf945 100755 --- a/compose.py +++ b/compose.py @@ -298,7 +298,7 @@ def create_argmap(images, skip_pull): dcgm_ver = re.search("DCGM_VERSION=([\S]{4,}) ", vars) dcgm_version = "" if dcgm_ver is None: - dcgm_version = "4.2.3-2" + dcgm_version = "4.4.0-1" log( "WARNING: DCGM version not found from image, installing the earlierst version {}".format( dcgm_version