vllm/.buildkite/test-pipeline.yaml

# In this file, you can add more tests to run either by adding a new step or
# adding a new command to an existing step. See different options here for examples.
# This script will be feed into Jinja template in `test-template.j2` to generate
# the final pipeline yaml file.

steps:
- label: Regression Test
  command: pytest -v -s test_regression.py
  working_dir: "/vllm-workspace/tests" # optional

- label: AsyncEngine Test
  command: pytest -v -s async_engine

- label: Basic Correctness Test
  commands:
  - VLLM_ATTENTION_BACKEND=XFORMERS pytest -v -s basic_correctness/test_basic_correctness.py
  - VLLM_ATTENTION_BACKEND=FLASH_ATTN pytest -v -s basic_correctness/test_basic_correctness.py
  - VLLM_ATTENTION_BACKEND=ROCM_FLASH pytest -v -s basic_correctness/test_basic_correctness.py
  - VLLM_ATTENTION_BACKEND=XFORMERS pytest -v -s basic_correctness/test_chunked_prefill.py
  - VLLM_ATTENTION_BACKEND=FLASH_ATTN pytest -v -s basic_correctness/test_chunked_prefill.py
  - VLLM_ATTENTION_BACKEND=ROCM_FLASH pytest -v -s basic_correctness/test_chunked_prefill.py

- label: Core Test
  command: pytest -v -s core

- label: Distributed Comm Ops Test
  command: pytest -v -s test_comm_ops.py
  working_dir: "/vllm-workspace/tests/distributed"
  num_gpus: 2 # only support 1 or 2 for now.

- label: Distributed Tests
  working_dir: "/vllm-workspace/tests/distributed"
  num_gpus: 2 # only support 1 or 2 for now.
  commands:
  - pytest -v -s test_pynccl.py
  - pytest -v -s test_pynccl_library.py
  - TEST_DIST_MODEL=facebook/opt-125m pytest -v -s test_basic_distributed_correctness.py
  - TEST_DIST_MODEL=meta-llama/Llama-2-7b-hf pytest -v -s test_basic_distributed_correctness.py
  - TEST_DIST_MODEL=facebook/opt-125m pytest -v -s test_chunked_prefill_distributed.py
  - TEST_DIST_MODEL=meta-llama/Llama-2-7b-hf pytest -v -s test_chunked_prefill_distributed.py

- label: Engine Test
  command: pytest -v -s engine tokenization test_sequence.py test_config.py

- label: Entrypoints Test
  commands:
  # these tests have to be separated, because each one will allocate all posible GPU memory
  - pytest -v -s entrypoints --ignore=entrypoints/test_server_oot_registration.py
  - pytest -v -s entrypoints/test_server_oot_registration.py

- label: Examples Test
  working_dir: "/vllm-workspace/examples"
  commands:
    # install aws cli for llava_example.py
    - pip install awscli
    - python3 offline_inference.py
    - python3 offline_inference_with_prefix.py
    - python3 llm_engine_example.py
    - python3 llava_example.py

- label: Kernels Test %N
  command: pytest -v -s kernels --shard-id=$$BUILDKITE_PARALLEL_JOB --num-shards=$$BUILDKITE_PARALLEL_JOB_COUNT
  parallelism: 4

- label: Models Test
  commands:
    - bash ../.buildkite/download-images.sh
    - pytest -v -s models --ignore=models/test_llava.py --ignore=models/test_mistral.py

- label: Llava Test
  commands:
    - bash ../.buildkite/download-images.sh
    - pytest -v -s models/test_llava.py

- label: Prefix Caching Test
  commands:
    - pytest -v -s prefix_caching

- label: Samplers Test
  command: pytest -v -s samplers

- label: LogitsProcessor Test
  command: pytest -v -s test_logits_processor.py

- label: Worker Test
  command: pytest -v -s worker

- label: Speculative decoding tests
  command: pytest -v -s spec_decode

- label: LoRA Test %N
  command: pytest -v -s lora --shard-id=$$BUILDKITE_PARALLEL_JOB --num-shards=$$BUILDKITE_PARALLEL_JOB_COUNT
  parallelism: 4

- label: Tensorizer Test
  command: apt-get install curl libsodium23 && pytest -v -s tensorizer_loader

- label: Metrics Test
  command: pytest -v -s metrics

- label: Benchmarks
  working_dir: "/vllm-workspace/.buildkite"
  commands:
  - pip install aiohttp
  - bash run-benchmarks.sh

- label: Documentation Build
  working_dir: "/vllm-workspace/test_docs/docs"
  no_gpu: True
  commands:
  - pip install -r requirements-docs.txt
  - SPHINXOPTS=\"-W\" make html