Feat: reimplement vllm backend beam search using logprobs #553
Workflow file for this run
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| name: CLI CUDA TensorRT-LLM Tests | |
| on: | |
| workflow_dispatch: | |
| push: | |
| branches: | |
| - main | |
| pull_request: | |
| branches: | |
| - main | |
| types: | |
| - opened | |
| - reopened | |
| - synchronize | |
| - labeled | |
| - unlabeled | |
| concurrency: | |
| cancel-in-progress: true | |
| group: ${{ github.workflow }}-${{ github.event.pull_request.number || github.ref }} | |
| jobs: | |
| cli_cuda_tensorrt_llm_single_gpu_tests: | |
| if: ${{ | |
| (github.event_name == 'push') || | |
| (github.event_name == 'workflow_dispatch') || | |
| contains( github.event.pull_request.labels.*.name, 'cli') || | |
| contains( github.event.pull_request.labels.*.name, 'cuda') || | |
| contains( github.event.pull_request.labels.*.name, 'tensorrt_llm') || | |
| contains( github.event.pull_request.labels.*.name, 'single_gpu') || | |
| contains( github.event.pull_request.labels.*.name, 'cli_cuda_tensorrt_llm') | |
| }} | |
| runs-on: | |
| group: aws-g5-4xlarge-plus | |
| container: | |
| image: huggingface/optimum-nvidia:latest | |
| options: --ipc host --gpus all | |
| steps: | |
| - name: Checkout | |
| uses: actions/checkout@v4 | |
| - name: Install dependencies | |
| run: | | |
| pip install -e .[testing] | |
| - name: Run tests | |
| run: | | |
| pytest tests/test_cli.py -x -s -k "cli and cuda and tensorrt_llm and not (tp or pp)" | |
| cli_cuda_tensorrt_llm_multi_gpu_tests: | |
| if: ${{ | |
| (github.event_name == 'push') || | |
| (github.event_name == 'workflow_dispatch') || | |
| contains( github.event.pull_request.labels.*.name, 'cli') || | |
| contains( github.event.pull_request.labels.*.name, 'cuda') || | |
| contains( github.event.pull_request.labels.*.name, 'tensorrt_llm') || | |
| contains( github.event.pull_request.labels.*.name, 'multi_gpu') || | |
| contains( github.event.pull_request.labels.*.name, 'cli_cuda_tensorrt_llm_multi_gpu') | |
| }} | |
| runs-on: | |
| group: aws-g5-12xlarge-plus | |
| container: | |
| image: huggingface/optimum-nvidia:latest | |
| options: --ipc host --gpus all | |
| steps: | |
| - name: Checkout | |
| uses: actions/checkout@v4 | |
| - name: Install dependencies | |
| run: | | |
| pip install -e .[testing] | |
| - name: Run tests (sequential) | |
| run: | | |
| FORCE_SEQUENTIAL=1 pytest tests/test_cli.py -x -s -k "cli and cuda and tensorrt_llm and (tp or pp)" |