Files
vllm/.buildkite/test_areas/models_language.yaml
2026-04-07 18:55:48 -07:00

101 lines
3.6 KiB
YAML

group: Models - Language
depends_on:
- image-build
steps:
- label: Language Models Tests (Standard)
timeout_in_minutes: 25
torch_nightly: true
source_file_dependencies:
- vllm/
- tests/models/language
commands:
# Test standard language models, excluding a subset of slow tests
- pip freeze | grep -E 'torch'
- pytest -v -s models/language -m 'core_model and (not slow_test)'
- label: Language Models Tests (Extra Standard) %N
timeout_in_minutes: 45
torch_nightly: true
source_file_dependencies:
- vllm/model_executor/models/
- tests/models/language/pooling/test_embedding.py
- tests/models/language/generation/test_common.py
- tests/models/language/pooling/test_classification.py
commands:
# Shard slow subset of standard language models tests. Only run when model
# source is modified, or when specified test files are modified
- pip freeze | grep -E 'torch'
- pytest -v -s models/language -m 'core_model and slow_test' --num-shards=$$BUILDKITE_PARALLEL_JOB_COUNT --shard-id=$$BUILDKITE_PARALLEL_JOB
parallelism: 2
- label: Language Models Tests (Hybrid) %N
timeout_in_minutes: 75
torch_nightly: true
source_file_dependencies:
- vllm/
- tests/models/language/generation
commands:
# Install fast path packages for testing against transformers
# Note: also needed to run plamo2 model in vLLM
- uv pip install --system --no-build-isolation 'git+https://github.com/state-spaces/mamba@v2.3.0'
- uv pip install --system --no-build-isolation 'git+https://github.com/Dao-AILab/causal-conv1d@v1.6.0'
# Shard hybrid language model tests
- pytest -v -s models/language/generation -m hybrid_model --num-shards=$$BUILDKITE_PARALLEL_JOB_COUNT --shard-id=$$BUILDKITE_PARALLEL_JOB
parallelism: 2
- label: Language Models Test (Extended Generation) # 80min
timeout_in_minutes: 110
optional: true
source_file_dependencies:
- vllm/
- tests/models/language/generation
commands:
# Install fast path packages for testing against transformers
# Note: also needed to run plamo2 model in vLLM
- uv pip install --system --no-build-isolation 'git+https://github.com/state-spaces/mamba@v2.3.0'
- uv pip install --system --no-build-isolation 'git+https://github.com/Dao-AILab/causal-conv1d@v1.6.0'
- pytest -v -s models/language/generation -m '(not core_model) and (not hybrid_model)'
mirror:
amd:
device: mi325_1
depends_on:
- image-build-amd
commands:
- uv pip install --system --no-build-isolation 'git+https://github.com/AndreasKaratzas/mamba@fix-rocm-7.0-warp-size-constexpr'
- uv pip install --system --no-build-isolation 'git+https://github.com/Dao-AILab/causal-conv1d@v1.5.2'
- pytest -v -s models/language/generation -m '(not core_model) and (not hybrid_model)'
- label: Language Models Test (PPL)
timeout_in_minutes: 110
device: h200_18gb
optional: true
source_file_dependencies:
- vllm/
- tests/models/language/generation_ppl_test
commands:
- pytest -v -s models/language/generation_ppl_test
- label: Language Models Test (Extended Pooling) # 36min
timeout_in_minutes: 50
optional: true
source_file_dependencies:
- vllm/
- tests/models/language/pooling
commands:
- pytest -v -s models/language/pooling -m 'not core_model'
mirror:
amd:
device: mi325_1
depends_on:
- image-build-amd
- label: Language Models Test (MTEB)
timeout_in_minutes: 110
device: h200_18gb
optional: true
source_file_dependencies:
- vllm/
- tests/models/language/pooling_mteb_test
commands:
- pytest -v -s models/language/pooling_mteb_test