Upgrade Dev containers for CICD to latest

Signed-off-by: Keval Morabia <28916987+kevalmorabia97@users.noreply.github.com>

Author: kevalmorabia97

.github/workflows/example_tests.yml

@@ -70,7 +70,7 @@ jobs:
     uses: ./.github/workflows/_example_tests_runner.yml
     secrets: inherit
     with:
-      docker_image: ${{ matrix.docker_image || 'nvcr.io/nvidia/pytorch:25.06-py3' }}
+      docker_image: ${{ matrix.docker_image || 'nvcr.io/nvidia/pytorch:26.01-py3' }}
       example: ${{ matrix.example }}
       pip_install_extras: "[hf,dev-test]"
       runner: linux-amd64-gpu-l4-latest-1
@@ -87,7 +87,7 @@ jobs:
     uses: ./.github/workflows/_example_tests_runner.yml
     secrets: inherit
     with:
-      docker_image: ${{ matrix.docker_image || 'nvcr.io/nvidia/pytorch:25.06-py3' }}
+      docker_image: ${{ matrix.docker_image || 'nvcr.io/nvidia/pytorch:26.01-py3' }}
       example: ${{ matrix.example }}
       pip_install_extras: "[hf,dev-test]"
       runner: linux-amd64-gpu-h100-latest-2
@@ -103,7 +103,7 @@ jobs:
     uses: ./.github/workflows/_example_tests_runner.yml
     secrets: inherit
     with:
-      docker_image: "nvcr.io/nvidia/tensorrt-llm/release:1.2.0rc4"
+      docker_image: "nvcr.io/nvidia/tensorrt-llm/release:1.2.0rc6.post3"
       example: ${{ matrix.example }}
       pip_install_extras: "[hf,dev-test]"
       runner: linux-amd64-gpu-h100-latest-1
@@ -117,7 +117,7 @@ jobs:
     uses: ./.github/workflows/_example_tests_runner.yml
     secrets: inherit
     with:
-      docker_image: "nvcr.io/nvidia/tensorrt-llm/release:1.2.0rc4"
+      docker_image: "nvcr.io/nvidia/tensorrt-llm/release:1.2.0rc6.post3"
       example: ${{ matrix.example }}
       pip_install_extras: "[hf,dev-test]"
       runner: linux-amd64-gpu-h100-latest-2
@@ -133,7 +133,7 @@ jobs:
     uses: ./.github/workflows/_example_tests_runner.yml
     secrets: inherit
     with:
-      docker_image: "nvcr.io/nvidia/tensorrt:25.08-py3"
+      docker_image: "nvcr.io/nvidia/tensorrt:26.01-py3"
       example: ${{ matrix.example }}
       pip_install_extras: "[all,dev-test]"
       runner: linux-amd64-gpu-l4-latest-1
@@ -147,7 +147,7 @@ jobs:
     uses: ./.github/workflows/_example_tests_runner.yml
     secrets: inherit
     with:
-      docker_image: "nvcr.io/nvidia/tensorrt:25.08-py3"
+      docker_image: "nvcr.io/nvidia/tensorrt:26.01-py3"
       example: ${{ matrix.example }}
       pip_install_extras: "[all,dev-test]"
       runner: linux-amd64-gpu-l4-latest-1

.github/workflows/gpu_tests.yml

@@ -63,14 +63,14 @@ jobs:
       fail-fast: false
       matrix:
         include:
-          - example: py312-cuda12-gpu
+          - example: py312-cuda13-gpu
             timeout: 90
-          - example: py312-cuda12-gpu-megatron
+          - example: py312-cuda13-gpu-megatron
             timeout: 120
     runs-on: linux-amd64-gpu-l4-latest-1
     timeout-minutes: ${{ matrix.timeout }}
     container: &gpu_container
-      image: nvcr.io/nvidia/pytorch:25.06-py3
+      image: nvcr.io/nvidia/pytorch:26.01-py3
       env:
         GIT_DEPTH: 1000 # For correct version for tests/gpu/torch/quantization/plugins/test_megatron.py
         PIP_CONSTRAINT: "" # Disable pip constraint for upgrading packages

.github/workflows/unit_tests.yml

@@ -37,7 +37,7 @@ jobs:
       - uses: actions/checkout@v6
       - uses: ./.github/actions/ubuntu-setup
       - name: Run unit tests
-        run: pip install tox && COV_ARGS="--cov" tox -e py312-torch29-tf_latest-unit
+        run: pip install tox && COV_ARGS="--cov" tox -e py312-torch210-tf_latest-unit
       - name: Upload coverage reports to Codecov
         uses: codecov/codecov-action@v5
         with:
@@ -55,7 +55,7 @@ jobs:
         with:
           python-version: "3.12"
       - name: Run unit tests (without coverage)
-        run: pip install tox && tox -e py312-torch29-tf_latest-unit
+        run: pip install tox && tox -e py312-torch210-tf_latest-unit
   multi-py:
     if: github.event_name == 'pull_request'
     needs: [linux]
@@ -70,15 +70,15 @@ jobs:
         with:
           python-version: "3.${{ matrix.py }}"
       - name: Run unit tests
-        run: pip install tox && tox -e py3${{ matrix.py }}-torch29-tf_latest-unit
+        run: pip install tox && tox -e py3${{ matrix.py }}-torch210-tf_latest-unit
   multi-torch:
     if: github.event_name == 'pull_request'
     needs: [linux]
     runs-on: ubuntu-latest
     timeout-minutes: 30
     strategy:
       matrix:
-        torch: [26, 27, 28]
+        torch: [26, 27, 28, 29]
     steps:
       - uses: actions/checkout@v6
       - uses: ./.github/actions/ubuntu-setup
@@ -96,7 +96,7 @@ jobs:
       - uses: actions/checkout@v6
       - uses: ./.github/actions/ubuntu-setup
       - name: Run unit tests
-        run: pip install tox && tox -e py312-torch29-tf_${{ matrix.tf }}-unit
+        run: pip install tox && tox -e py312-torch210-tf_${{ matrix.tf }}-unit
   partial-install:
     if: github.event_name == 'pull_request'
     needs: [linux]

docs/source/getting_started/_installation_for_Linux.rst

@@ -14,11 +14,11 @@ Latest Model Optimizer (``nvidia-modelopt``) currently has the following system
 +-------------------------+-----------------------------+
 | Python                  |  >=3.10,<3.13               |
 +-------------------------+-----------------------------+
-| CUDA                    |  >=12.0                     |
+| CUDA                    |  12.x, 13.x                 |
 +-------------------------+-----------------------------+
 | PyTorch                 |  >=2.6                      |
 +-------------------------+-----------------------------+
-| TensorRT-LLM (Optional) |  1.2.0rc4                   |
+| TensorRT-LLM (Optional) |  >=1.0                      |
 +-------------------------+-----------------------------+
 | ONNX Runtime (Optional) |  1.22                       |
 +-------------------------+-----------------------------+
@@ -126,6 +126,10 @@ Additionally, we support installing dependencies for following 3rd-party package
     *   - Huggingface (``transformers``, ``diffusers``, etc.)
         - ``[hf]``
 
+**CUDA specific dependencies**
+
+* By default, ``cupy-cuda12x`` is installed for INT4 ONNX quantization. If you have CUDA 13, you need to run ``pip uninstall -y cupy-cuda12x`` and ``pip install cupy-cuda13x`` after installing ``nvidia-modelopt[onnx]``.
+
 **Accelerated Quantization with Triton Kernels**
 
 ModelOpt includes optimized quantization kernels implemented with Triton language that accelerate quantization

tox.ini

@@ -1,22 +1,24 @@
 [tox]
 envlist=
     pre-commit-all
-    py312-torch28-tf_latest-unit
-    py312-cuda12-gpu
+    py312-torch210-tf_latest-unit
+    py312-cuda13-gpu
+    py312-cuda13-gpu-megatron
 skipsdist = True
 toxworkdir = /tmp/{env:USER}-modelopt-tox
 
 
 ############################
 # CPU Unit test environments
 ############################
-[testenv:{py310,py311,py312}-torch{26,27,28,29}-tf_{min,latest}-unit]
+[testenv:{py310,py311,py312}-torch{26,27,28,29,210}-tf_{min,latest}-unit]
 deps =
     # torch version auto-selected based on torchvision version
     torch26: torchvision~=0.21.0
     torch27: torchvision~=0.22.0
     torch28: torchvision~=0.23.0
     torch29: torchvision~=0.24.0
+    torch210: torchvision~=0.25.0
 
     # Install megatron-core for special unit tests
     megatron-core
@@ -36,8 +38,8 @@ commands =
 allowlist_externals =
     bash, rm
 deps =
-    # Make sure torch 2.9 is used
-    torchvision~=0.24.0
+    # Make sure torch 2.10 is used
+    torchvision~=0.25.0
 
     # ONNX unit tests heavily rely on torch / torchvision
     onnx: .[onnx,dev-test]
@@ -57,7 +59,7 @@ commands =
 ###########################################################
 # GPU test environments (Should be used with --current-env)
 ###########################################################
-[testenv:{py310,py311,py312}-cuda12-gpu]
+[testenv:{py310,py311,py312}-cuda13-gpu]
 commands_pre =
     # Install deps here so that it gets installed even in --current-env
     pip install git+https://github.com/Dao-AILab/fast-hadamard-transform.git
@@ -67,7 +69,7 @@ commands =
     # Coverage fails with "Can't combine line data with arc data" error so not using "--cov"
     python -m pytest tests/gpu
 
-[testenv:{py310,py311,py312}-cuda12-gpu-megatron]
+[testenv:{py310,py311,py312}-cuda13-gpu-megatron]
 commands_pre =
     # Install deps here so that it gets installed even in --current-env
     pip install -U megatron-core