diff --git a/Dockerfile b/Dockerfile index 37f57605..0ad0045b 100644 --- a/Dockerfile +++ b/Dockerfile @@ -1,8 +1,8 @@ # SPDX-FileCopyrightText: Copyright (c) 2020-2026 NVIDIA CORPORATION & AFFILIATES. All rights reserved. # SPDX-License-Identifier: Apache-2.0 -ARG BASE_IMAGE=nvcr.io/nvidia/tritonserver:26.03-py3 -ARG TRITONSDK_BASE_IMAGE=nvcr.io/nvidia/tritonserver:26.03-py3-sdk +ARG BASE_IMAGE=nvcr.io/nvidia/tritonserver:26.04-py3 +ARG TRITONSDK_BASE_IMAGE=nvcr.io/nvidia/tritonserver:26.04-py3-sdk ARG MODEL_ANALYZER_VERSION=1.54.0dev ARG MODEL_ANALYZER_CONTAINER_VERSION=26.05dev @@ -20,9 +20,15 @@ RUN apt update -qq && apt install -y docker.io wkhtmltopdf # Install tritonclient COPY --from=sdk /workspace/install/python /tmp/tritonclient -RUN find /tmp/tritonclient -maxdepth 1 -type f -name \ - "tritonclient-*-manylinux*.whl" | xargs printf -- '%s[all]' | \ - xargs pip3 install --upgrade && rm -rf /tmp/tritonclient/ + +RUN --mount=type=secret,id=triton_ci_pip_extra_values,env=TRITON_CI_PYPI_EXTRA_VALUES \ + if [ -n "${TRITON_CI_PYPI_EXTRA_VALUES}" ]; then \ + find /tmp/tritonclient -maxdepth 1 -type f -name \ + "tritonclient-*-any*.whl" -exec pip3 install --upgrade ${TRITON_CI_PYPI_EXTRA_VALUES} {}[all] \; ; \ + else \ + find /tmp/tritonclient -maxdepth 1 -type f -name \ + "tritonclient-*-any*.whl" -exec pip3 install --upgrade {}[all] \; ; \ + fi WORKDIR /opt/triton-model-analyzer diff --git a/README.md b/README.md index 22d1f15d..f715cebc 100644 --- a/README.md +++ b/README.md @@ -17,14 +17,14 @@ Triton Model Analyzer is a CLI tool which can help you find a more optimal confi - [Optuna Search](docs/config_search.md#optuna-search-mode) **_-ALPHA RELEASE-_** allows you to search for every parameter that can be specified in the model configuration, using a hyperparameter optimization framework. Please see the [Optuna](https://optuna.org/) website if you are interested in specific details on how the algorithm functions. -- [Quick Search](docs/config_search.md#quick-search-mode) will **sparsely** search the [Max Batch Size](https://github.com/triton-inference-server/server/blob/r26.03/docs/user_guide/model_configuration.md#maximum-batch-size), - [Dynamic Batching](https://github.com/triton-inference-server/server/blob/r26.03/docs/user_guide/batcher.md#dynamic-batcher), and - [Instance Group](https://github.com/triton-inference-server/server/blob/r26.03/docs/user_guide/model_configuration.md#instance-groups) spaces by utilizing a heuristic hill-climbing algorithm to help you quickly find a more optimal configuration +- [Quick Search](docs/config_search.md#quick-search-mode) will **sparsely** search the [Max Batch Size](https://github.com/triton-inference-server/server/blob/r26.04/docs/user_guide/model_configuration.md#maximum-batch-size), + [Dynamic Batching](https://github.com/triton-inference-server/server/blob/r26.04/docs/user_guide/batcher.md#dynamic-batcher), and + [Instance Group](https://github.com/triton-inference-server/server/blob/r26.04/docs/user_guide/model_configuration.md#instance-groups) spaces by utilizing a heuristic hill-climbing algorithm to help you quickly find a more optimal configuration - [Automatic Brute Search](docs/config_search.md#automatic-brute-search) will **exhaustively** search the - [Max Batch Size](https://github.com/triton-inference-server/server/blob/r26.03/docs/user_guide/model_configuration.md#maximum-batch-size), - [Dynamic Batching](https://github.com/triton-inference-server/server/blob/r26.03/docs/user_guide/batcher.md#dynamic-batcher), and - [Instance Group](https://github.com/triton-inference-server/server/blob/r26.03/docs/user_guide/model_configuration.md#instance-groups) + [Max Batch Size](https://github.com/triton-inference-server/server/blob/r26.04/docs/user_guide/model_configuration.md#maximum-batch-size), + [Dynamic Batching](https://github.com/triton-inference-server/server/blob/r26.04/docs/user_guide/batcher.md#dynamic-batcher), and + [Instance Group](https://github.com/triton-inference-server/server/blob/r26.04/docs/user_guide/model_configuration.md#instance-groups) parameters of your model configuration - [Manual Brute Search](docs/config_search.md#manual-brute-search) allows you to create manual sweeps for every parameter that can be specified in the model configuration diff --git a/docs/bls_quick_start.md b/docs/bls_quick_start.md index a7eabf07..fd0a1848 100644 --- a/docs/bls_quick_start.md +++ b/docs/bls_quick_start.md @@ -38,7 +38,7 @@ git pull origin main **1. Pull the SDK container:** ``` -docker pull nvcr.io/nvidia/tritonserver:26.03-py3-sdk +docker pull nvcr.io/nvidia/tritonserver:26.04-py3-sdk ``` **2. Run the SDK container** @@ -48,7 +48,7 @@ docker run -it --gpus 1 \ --shm-size 2G \ -v /var/run/docker.sock:/var/run/docker.sock \ -v $(pwd)/examples/quick-start:$(pwd)/examples/quick-start \ - --net=host nvcr.io/nvidia/tritonserver:26.03-py3-sdk + --net=host nvcr.io/nvidia/tritonserver:26.04-py3-sdk ``` **Important:** The example above uses a single GPU. If you are running on multiple GPUs, you may need to increase the shared memory size accordingly

@@ -57,7 +57,7 @@ docker run -it --gpus 1 \ --- -The [examples/quick-start](../examples/quick-start) directory is an example [Triton Model Repository](https://github.com/triton-inference-server/server/blob/r26.03/docs/user_guide/model_repository.md) that contains the BLS model `bls` which calculates the sum of two inputs using `add` model. +The [examples/quick-start](../examples/quick-start) directory is an example [Triton Model Repository](https://github.com/triton-inference-server/server/blob/r26.04/docs/user_guide/model_repository.md) that contains the BLS model `bls` which calculates the sum of two inputs using `add` model. An example model analyzer YAML config that performs a BLS model search diff --git a/docs/config.md b/docs/config.md index e4d7d7b9..de3e3b4c 100644 --- a/docs/config.md +++ b/docs/config.md @@ -142,7 +142,7 @@ cpu_only_composing_models: [ reload_model_disable: | default: false] # Triton Docker image tag used when launching using Docker mode -[ triton_docker_image: | default: nvcr.io/nvidia/tritonserver:26.03-py3 ] +[ triton_docker_image: | default: nvcr.io/nvidia/tritonserver:26.04-py3 ] # Triton Server HTTP endpoint url used by Model Analyzer client" [ triton_http_endpoint: | default: localhost:8000 ] diff --git a/docs/ensemble_quick_start.md b/docs/ensemble_quick_start.md index 9ba7ef81..e74189c4 100644 --- a/docs/ensemble_quick_start.md +++ b/docs/ensemble_quick_start.md @@ -38,7 +38,7 @@ git pull origin main **1. Pull the SDK container:** ``` -docker pull nvcr.io/nvidia/tritonserver:26.03-py3-sdk +docker pull nvcr.io/nvidia/tritonserver:26.04-py3-sdk ``` **2. Run the SDK container** @@ -48,7 +48,7 @@ docker run -it --gpus 1 \ --shm-size 1G \ -v /var/run/docker.sock:/var/run/docker.sock \ -v $(pwd)/examples/quick-start:$(pwd)/examples/quick-start \ - --net=host nvcr.io/nvidia/tritonserver:26.03-py3-sdk + --net=host nvcr.io/nvidia/tritonserver:26.04-py3-sdk ``` **Important:** The example above uses a single GPU. If you are running on multiple GPUs, you may need to increase the shared memory size accordingly

@@ -57,7 +57,7 @@ docker run -it --gpus 1 \ --- -The [examples/quick-start](../examples/quick-start) directory is an example [Triton Model Repository](https://github.com/triton-inference-server/server/blob/r26.03/docs/user_guide/model_repository.md) that contains the ensemble model `ensemble_add_sub`, which calculates the sum and difference of two inputs using `add` and `sub` models. +The [examples/quick-start](../examples/quick-start) directory is an example [Triton Model Repository](https://github.com/triton-inference-server/server/blob/r26.04/docs/user_guide/model_repository.md) that contains the ensemble model `ensemble_add_sub`, which calculates the sum and difference of two inputs using `add` and `sub` models. Run the Model Analyzer `profile` subcommand inside the container with: diff --git a/docs/kubernetes_deploy.md b/docs/kubernetes_deploy.md index bea98bee..d4bafa5b 100644 --- a/docs/kubernetes_deploy.md +++ b/docs/kubernetes_deploy.md @@ -68,7 +68,7 @@ images: triton: image: nvcr.io/nvidia/tritonserver - tag: 26.03-py3 + tag: 26.04-py3 ``` The model analyzer executable uses the config file defined in `helm-chart/templates/config-map.yaml`. This config can be modified to supply arguments to model analyzer. Only the content under the `config.yaml` section of the file should be modified. diff --git a/docs/mm_quick_start.md b/docs/mm_quick_start.md index b5bf0d2e..8100c3d4 100644 --- a/docs/mm_quick_start.md +++ b/docs/mm_quick_start.md @@ -38,7 +38,7 @@ git pull origin main **1. Pull the SDK container:** ``` -docker pull nvcr.io/nvidia/tritonserver:26.03-py3-sdk +docker pull nvcr.io/nvidia/tritonserver:26.04-py3-sdk ``` **2. Run the SDK container** @@ -47,7 +47,7 @@ docker pull nvcr.io/nvidia/tritonserver:26.03-py3-sdk docker run -it --gpus all \ -v /var/run/docker.sock:/var/run/docker.sock \ -v $(pwd)/examples/quick-start:$(pwd)/examples/quick-start \ - --net=host nvcr.io/nvidia/tritonserver:26.03-py3-sdk + --net=host nvcr.io/nvidia/tritonserver:26.04-py3-sdk ``` ## `Step 3:` Profile both models concurrently @@ -55,7 +55,7 @@ docker run -it --gpus all \ --- The [examples/quick-start](../examples/quick-start) directory is an example -[Triton Model Repository](https://github.com/triton-inference-server/server/blob/r26.03/docs/user_guide/model_repository.md) that contains two libtorch models: `add_sub` & `resnet50_python` +[Triton Model Repository](https://github.com/triton-inference-server/server/blob/r26.04/docs/user_guide/model_repository.md) that contains two libtorch models: `add_sub` & `resnet50_python` Run the Model Analyzer `profile` subcommand inside the container with: diff --git a/docs/quick_start.md b/docs/quick_start.md index fe8508bb..30625344 100644 --- a/docs/quick_start.md +++ b/docs/quick_start.md @@ -38,7 +38,7 @@ git pull origin main **1. Pull the SDK container:** ``` -docker pull nvcr.io/nvidia/tritonserver:26.03-py3-sdk +docker pull nvcr.io/nvidia/tritonserver:26.04-py3-sdk ``` **2. Run the SDK container** @@ -47,7 +47,7 @@ docker pull nvcr.io/nvidia/tritonserver:26.03-py3-sdk docker run -it --gpus all \ -v /var/run/docker.sock:/var/run/docker.sock \ -v $(pwd)/examples/quick-start:$(pwd)/examples/quick-start \ - --net=host nvcr.io/nvidia/tritonserver:26.03-py3-sdk + --net=host nvcr.io/nvidia/tritonserver:26.04-py3-sdk ``` ## `Step 3:` Profile the `add_sub` model @@ -55,7 +55,7 @@ docker run -it --gpus all \ --- The [examples/quick-start](../examples/quick-start) directory is an example -[Triton Model Repository](https://github.com/triton-inference-server/server/blob/r26.03/docs/user_guide/model_repository.md) that contains a simple libtorch model which calculates +[Triton Model Repository](https://github.com/triton-inference-server/server/blob/r26.04/docs/user_guide/model_repository.md) that contains a simple libtorch model which calculates the sum and difference of two inputs. Run the Model Analyzer `profile` subcommand inside the container with: diff --git a/helm-chart/values.yaml b/helm-chart/values.yaml index 7f7583ea..0c7d687e 100644 --- a/helm-chart/values.yaml +++ b/helm-chart/values.yaml @@ -26,4 +26,4 @@ images: triton: image: nvcr.io/nvidia/tritonserver - tag: 26.03-py3 + tag: 26.04-py3 diff --git a/model_analyzer/config/input/config_defaults.py b/model_analyzer/config/input/config_defaults.py index e7c8cff2..999cbed5 100755 --- a/model_analyzer/config/input/config_defaults.py +++ b/model_analyzer/config/input/config_defaults.py @@ -52,7 +52,7 @@ DEFAULT_CONCURRENCY_SWEEP_DISABLE = False DEFAULT_DCGM_DISABLE = False DEFAULT_TRITON_LAUNCH_MODE = "local" -DEFAULT_TRITON_DOCKER_IMAGE = "nvcr.io/nvidia/tritonserver:26.03-py3" +DEFAULT_TRITON_DOCKER_IMAGE = "nvcr.io/nvidia/tritonserver:26.04-py3" DEFAULT_TRITON_HTTP_ENDPOINT = "localhost:8000" DEFAULT_TRITON_GRPC_ENDPOINT = "localhost:8001" DEFAULT_TRITON_METRICS_URL = "http://localhost:8002/metrics"