Model-Optimizer/.github/workflows/example_tests.yml at a4c2408cd4acb245e6cb25eaa493ba429c774a14 · NVIDIA/Model-Optimizer · GitHub

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
name: Example tests

on:
  push:
    branches: ["pull-request/[0-9]+"]
    # NOTE: paths cannot be used since push happens to copied PR and only latest commit to PR is used
  schedule:
    - cron: "0 0 * * *" # Nightly
  workflow_dispatch:
    # On-demand


concurrency:
  # Cancel previous runs if new commit is pushed to the same PR
  group: ${{ github.workflow }}-${{ startsWith(github.ref, 'refs/heads/pull-request/') && github.ref || github.sha }}
  cancel-in-progress: true

jobs:
  pr-gate:
    uses: ./.github/workflows/_pr_gate.yml
    permissions:
      checks: read
    secrets: inherit
    with:
      files: |
        .github/workflows/example_tests.yml
        examples/**
        modelopt/**
        pyproject.toml
        tests/examples/**

  ##### PyTorch Example Tests (speculative_decoding requires 26.01 image) #####
  torch:
    needs: [pr-gate]
    if: >-
      needs.pr-gate.outputs.any_changed == 'true' ||
      !startsWith(github.ref, 'refs/heads/pull-request/')
    strategy:
      fail-fast: false
      matrix:
        example: [llm_distill, llm_qat, llm_sparsity, diffusers_sparsity]
        include:
          - example: speculative_decoding
            docker_image: "26.01"
    uses: ./.github/workflows/_example_tests_runner.yml
    secrets: inherit
    with:
      docker_image: "nvcr.io/nvidia/pytorch:${{ matrix.docker_image || '26.03' }}-py3"
      example: ${{ matrix.example }}
      timeout_minutes: 30
      pip_install_extras: "[hf,dev-test]"
      runner: ${{ startsWith(github.ref, 'refs/heads/pull-request/') && 'linux-amd64-gpu-rtxpro6000-latest-1' || 'linux-amd64-gpu-rtxpro6000-latest-2' }}

  ##### TensorRT-LLM Example Tests (pr/non-pr split: non-pr runs extra autodeploy+eval examples) #####
  trtllm-pr:
    needs: [pr-gate]
    if: startsWith(github.ref, 'refs/heads/pull-request/') && needs.pr-gate.outputs.any_changed == 'true'
    strategy:
      fail-fast: false
      matrix:
        example: [llm_ptq, vlm_ptq]
    uses: ./.github/workflows/_example_tests_runner.yml
    secrets: inherit
    with:
      docker_image: "nvcr.io/nvidia/tensorrt-llm/release:1.3.0rc10"
      example: ${{ matrix.example }}
      pip_install_extras: "[hf,dev-test]"
      runner: linux-amd64-gpu-rtxpro6000-latest-1

  trtllm-non-pr:
    if: ${{ !startsWith(github.ref, 'refs/heads/pull-request/') }}
    strategy:
      fail-fast: false
      matrix:
        example: [llm_autodeploy, llm_eval, llm_ptq, vlm_ptq]
    uses: ./.github/workflows/_example_tests_runner.yml
    secrets: inherit
    with:
      docker_image: "nvcr.io/nvidia/tensorrt-llm/release:1.3.0rc10"
      example: ${{ matrix.example }}
      pip_install_extras: "[hf,dev-test]"
      runner: linux-amd64-gpu-rtxpro6000-latest-2

  ##### Megatron Example Tests #####
  megatron:
    needs: [pr-gate]
    if: >-
      needs.pr-gate.outputs.any_changed == 'true' ||
      !startsWith(github.ref, 'refs/heads/pull-request/')
    uses: ./.github/workflows/_example_tests_runner.yml
    secrets: inherit
    with:
      docker_image: "nvcr.io/nvidia/nemo:26.02"
      example: megatron_bridge
      timeout_minutes: 30
      pip_install_extras: "[hf,puzzletron,dev-test]"
      runner: ${{ startsWith(github.ref, 'refs/heads/pull-request/') && 'linux-amd64-gpu-rtxpro6000-latest-1' || 'linux-amd64-gpu-rtxpro6000-latest-2' }}

  ##### ONNX/TensorRT Example Tests #####
  onnx:
    needs: [pr-gate]
    if: >-
      needs.pr-gate.outputs.any_changed == 'true' ||
      !startsWith(github.ref, 'refs/heads/pull-request/')
    strategy:
      fail-fast: false
      matrix:
        example: [diffusers, torch_onnx]
    uses: ./.github/workflows/_example_tests_runner.yml
    secrets: inherit
    with:
      docker_image: "nvcr.io/nvidia/tensorrt:26.02-py3"
      example: ${{ matrix.example }}
      pip_install_extras: "[onnx,hf,dev-test]"
      runner: ${{ startsWith(github.ref, 'refs/heads/pull-request/') && 'linux-amd64-gpu-rtxpro6000-latest-1' || 'linux-amd64-gpu-rtxpro6000-latest-2' }}

  ##### Required Check for PR #####
  example-pr-required-check:
    # Run even if example tests are skipped
    if: ${{ startsWith(github.ref, 'refs/heads/pull-request/') && always() }}
    needs: [pr-gate, torch, trtllm-pr, megatron, onnx]
    runs-on: ubuntu-latest
    steps:
      - name: Required example tests did not succeed
        if: |
          needs.pr-gate.result != 'success' ||
          (needs.pr-gate.outputs.any_changed == 'true' && (
            needs.torch.result != 'success' ||
            needs.trtllm-pr.result != 'success' ||
            needs.megatron.result != 'success' ||
            needs.onnx.result != 'success'
          ))
        run: exit 1