From d6fed669b83a00002ae6d12e14021b4709556e2f Mon Sep 17 00:00:00 2001 From: Misha Chornyi <99709299+mc-nv@users.noreply.github.com> Date: Wed, 13 May 2026 13:50:05 -0700 Subject: [PATCH 1/2] docs(versions): Update README and versions for r26.05 (#1023) VERSION 1.54.0dev -> 1.54.0. Dockerfile MODEL_ANALYZER_VERSION and MODEL_ANALYZER_CONTAINER_VERSION ARGs drop the dev suffix. Bump nvcr.io/nvidia/tritonserver tag references from 26.04-py3* to 26.05-py3* and r26.04 server doc links to r26.05 across README, helm-chart/values.yaml, docs/{bls,ensemble,mm,quick}_quick_start.md, docs/config.md, docs/kubernetes_deploy.md, and model_analyzer/config/input/config_defaults.py. Refs TRI-1048 --- Dockerfile | 8 ++++---- README.md | 12 ++++++------ VERSION | 2 +- docs/bls_quick_start.md | 6 +++--- docs/config.md | 2 +- docs/ensemble_quick_start.md | 6 +++--- docs/kubernetes_deploy.md | 2 +- docs/mm_quick_start.md | 6 +++--- docs/quick_start.md | 6 +++--- helm-chart/values.yaml | 2 +- model_analyzer/config/input/config_defaults.py | 2 +- 11 files changed, 27 insertions(+), 27 deletions(-) diff --git a/Dockerfile b/Dockerfile index 0ad0045b..c94de8a5 100644 --- a/Dockerfile +++ b/Dockerfile @@ -1,11 +1,11 @@ # 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:26.04-py3 -ARG TRITONSDK_BASE_IMAGE=nvcr.io/nvidia/tritonserver:26.04-py3-sdk +ARG BASE_IMAGE=nvcr.io/nvidia/tritonserver:26.05-py3 +ARG TRITONSDK_BASE_IMAGE=nvcr.io/nvidia/tritonserver:26.05-py3-sdk -ARG MODEL_ANALYZER_VERSION=1.54.0dev -ARG MODEL_ANALYZER_CONTAINER_VERSION=26.05dev +ARG MODEL_ANALYZER_VERSION=1.54.0 +ARG MODEL_ANALYZER_CONTAINER_VERSION=26.05 FROM ${TRITONSDK_BASE_IMAGE} AS sdk FROM ${BASE_IMAGE} diff --git a/README.md b/README.md index f715cebc..19ec788d 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/r26.04/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 +- [Quick Search](docs/config_search.md#quick-search-mode) will **sparsely** search the [Max Batch Size](https://github.com/triton-inference-server/server/blob/r26.05/docs/user_guide/model_configuration.md#maximum-batch-size), + [Dynamic Batching](https://github.com/triton-inference-server/server/blob/r26.05/docs/user_guide/batcher.md#dynamic-batcher), and + [Instance Group](https://github.com/triton-inference-server/server/blob/r26.05/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/r26.04/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) + [Max Batch Size](https://github.com/triton-inference-server/server/blob/r26.05/docs/user_guide/model_configuration.md#maximum-batch-size), + [Dynamic Batching](https://github.com/triton-inference-server/server/blob/r26.05/docs/user_guide/batcher.md#dynamic-batcher), and + [Instance Group](https://github.com/triton-inference-server/server/blob/r26.05/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/VERSION b/VERSION index e705dccb..b7921ae8 100644 --- a/VERSION +++ b/VERSION @@ -1 +1 @@ -1.54.0dev +1.54.0 diff --git a/docs/bls_quick_start.md b/docs/bls_quick_start.md index fd0a1848..b3ff19a2 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:26.04-py3-sdk +docker pull nvcr.io/nvidia/tritonserver:26.05-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:26.04-py3-sdk + --net=host nvcr.io/nvidia/tritonserver:26.05-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/r26.04/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/r26.05/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 de3e3b4c..5f861a67 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:26.04-py3 ] +[ triton_docker_image: | default: nvcr.io/nvidia/tritonserver:26.05-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 e74189c4..57eb03cc 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:26.04-py3-sdk +docker pull nvcr.io/nvidia/tritonserver:26.05-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:26.04-py3-sdk + --net=host nvcr.io/nvidia/tritonserver:26.05-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/r26.04/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. +The [examples/quick-start](../examples/quick-start) directory is an example [Triton Model Repository](https://github.com/triton-inference-server/server/blob/r26.05/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 d4bafa5b..297a5866 100644 --- a/docs/kubernetes_deploy.md +++ b/docs/kubernetes_deploy.md @@ -68,7 +68,7 @@ images: triton: image: nvcr.io/nvidia/tritonserver - tag: 26.04-py3 + tag: 26.05-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 8100c3d4..35d4d266 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:26.04-py3-sdk +docker pull nvcr.io/nvidia/tritonserver:26.05-py3-sdk ``` **2. Run the SDK container** @@ -47,7 +47,7 @@ docker pull nvcr.io/nvidia/tritonserver:26.04-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:26.04-py3-sdk + --net=host nvcr.io/nvidia/tritonserver:26.05-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/r26.04/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/r26.05/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 30625344..320a3ed4 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:26.04-py3-sdk +docker pull nvcr.io/nvidia/tritonserver:26.05-py3-sdk ``` **2. Run the SDK container** @@ -47,7 +47,7 @@ docker pull nvcr.io/nvidia/tritonserver:26.04-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:26.04-py3-sdk + --net=host nvcr.io/nvidia/tritonserver:26.05-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/r26.04/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/r26.05/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 0c7d687e..2195476f 100644 --- a/helm-chart/values.yaml +++ b/helm-chart/values.yaml @@ -26,4 +26,4 @@ images: triton: image: nvcr.io/nvidia/tritonserver - tag: 26.04-py3 + tag: 26.05-py3 diff --git a/model_analyzer/config/input/config_defaults.py b/model_analyzer/config/input/config_defaults.py index 999cbed5..feac2dd6 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:26.04-py3" +DEFAULT_TRITON_DOCKER_IMAGE = "nvcr.io/nvidia/tritonserver:26.05-py3" DEFAULT_TRITON_HTTP_ENDPOINT = "localhost:8000" DEFAULT_TRITON_GRPC_ENDPOINT = "localhost:8001" DEFAULT_TRITON_METRICS_URL = "http://localhost:8002/metrics" From fdc036fa2595c832de160f778430f49e077bcd3d Mon Sep 17 00:00:00 2001 From: "M. Chornyi" <99709299+mc-nv@users.noreply.github.com> Date: Wed, 27 May 2026 12:37:19 -0700 Subject: [PATCH 2/2] fix: Update REAMDE.md --- README.md | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-) diff --git a/README.md b/README.md index 19ec788d..8a0ee4ed 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/r26.05/docs/user_guide/model_configuration.md#maximum-batch-size), - [Dynamic Batching](https://github.com/triton-inference-server/server/blob/r26.05/docs/user_guide/batcher.md#dynamic-batcher), and - [Instance Group](https://github.com/triton-inference-server/server/blob/r26.05/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/main/docs/user_guide/model_configuration.md#maximum-batch-size), + [Dynamic Batching](https://github.com/triton-inference-server/server/blob/main/docs/user_guide/batcher.md#dynamic-batcher), and + [Instance Group](https://github.com/triton-inference-server/server/blob/main/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/r26.05/docs/user_guide/model_configuration.md#maximum-batch-size), - [Dynamic Batching](https://github.com/triton-inference-server/server/blob/r26.05/docs/user_guide/batcher.md#dynamic-batcher), and - [Instance Group](https://github.com/triton-inference-server/server/blob/r26.05/docs/user_guide/model_configuration.md#instance-groups) + [Max Batch Size](https://github.com/triton-inference-server/server/blob/main/docs/user_guide/model_configuration.md#maximum-batch-size), + [Dynamic Batching](https://github.com/triton-inference-server/server/blob/main/docs/user_guide/batcher.md#dynamic-batcher), and + [Instance Group](https://github.com/triton-inference-server/server/blob/main/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