6000 lines
223 KiB
YAML
6000 lines
223 KiB
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-amd.j2` at
|
|
# https://github.com/vllm-project/buildkite-ci/blob/main/scripts/test-template-amd.j2
|
|
# to generate the final pipeline yaml file.
|
|
|
|
# Documentation
|
|
# label(str): the name of the test. emojis allowed.
|
|
# fast_check(bool): whether to run this on each commit on the fastcheck pipeline.
|
|
# fast_check_only(bool): run this test on the fastcheck pipeline only
|
|
# optional(bool): never run this test by default (i.e. need to unblock manually) unless it's a scheduled nightly run.
|
|
# soft_fail(bool): allow this step to fail without failing the entire pipeline (useful for flaky or experimental tests).
|
|
# command(str): the single command to run for tests. incompatible with commands.
|
|
# commands(list): the list of commands to run for the test. incompatible with command.
|
|
# mirror_hardwares(list): selector tags for AMD pipelines that should include the test.
|
|
# dind(bool): when false, run the job directly in the AMD Kubernetes pod.
|
|
# When true or omitted, use the legacy Docker-in-Docker path.
|
|
# num_gpus(int): override the number of GPUs for the test. defaults to 1 GPU. currently supports 2,4.
|
|
# num_nodes(int): whether to simulate multi-node setup by launching multiple containers on one host,
|
|
# in this case, commands must be specified. the first command runs on the first host, the second
|
|
# command runs on the second host.
|
|
# timeout_in_minutes(int): sets a timeout for the step in minutes. if not specified, uses the default timeout.
|
|
# parallelism(int): number of parallel jobs to run for this step. enables test sharding using $$BUILDKITE_PARALLEL_JOB
|
|
# and $$BUILDKITE_PARALLEL_JOB_COUNT environment variables.
|
|
# working_dir(str): specify the place where the command should execute, default to /vllm-workspace/tests
|
|
# source_file_dependencies(list): the list of prefixes to opt-in the test for, if empty, the test will always run.
|
|
|
|
# When adding a test
|
|
# - If the test belongs to an existing group, add it there
|
|
# - If the test is short, add to any existing step
|
|
# - If the test takes more than 10min, then it is okay to create a new step.
|
|
# Note that all steps execute in parallel.
|
|
|
|
|
|
#####################################################################################################################################
|
|
# #
|
|
# README #
|
|
# #
|
|
#####################################################################################################################################
|
|
# #
|
|
# IMPORTANT: #
|
|
# * Currently AMD CI has MI250 agents, MI300 agents, and MI355 agents. All upcoming feature improvements are #
|
|
# tracked in: https://github.com/vllm-project/vllm/issues/34994 #
|
|
# #
|
|
#-----------------------------------------------------------------------------------------------------------------------------------#
|
|
# #
|
|
# NOTES: #
|
|
# * [Pytorch Nightly Dependency Override Check]: if this test fails, it means the nightly torch version is not compatible with #
|
|
# some of the dependencies. Please check the error message and add the package to #
|
|
# whitelist in `/vllm/tools/pre_commit/generate_nightly_torch_test.py`. #
|
|
# * [Entrypoints Integration (LLM)]: #
|
|
# - {`pytest -v -s entrypoints/llm/test_generate.py`}: It needs a clean process #
|
|
# - {`pytest -v -s entrypoints/offline_mode`}: Needs to avoid interference with other tests #
|
|
# * [Engine / Engine (1 GPU) / e2e Scheduling / e2e Core / V1 e2e / Spec Decode / V1 Sample + Logits / V1 Core + KV + Metrics]: #
|
|
# - Previously a single "V1 Test e2e + engine" step, now split across multiple groups. #
|
|
# - V1 e2e (2/4 GPUs) uses 4 GPUs but is scheduled on 8-GPU machines for stability. See: #
|
|
# https://github.com/vllm-project/vllm/pull/31040 #
|
|
# * [V1 Sample + Logits / V1 Core + KV + Metrics / V1 others (CPU)]: #
|
|
# - Previously a single "V1 others" step, now split to avoid interference. #
|
|
# - Integration test for streaming correctness (requires special branch for __harness__ lib). #
|
|
# * [V1 others (CPU)]: Split the tests to avoid interference #
|
|
# * [PyTorch Compilation Unit Tests]: Run unit tests defined directly under `compile/`, not including subdirectories, which #
|
|
# are usually heavier tests covered elsewhere. Use `find` to launch multiple instances #
|
|
# of pytest so that they do not suffer from: #
|
|
# https://github.com/vllm-project/vllm/issues/28965 #
|
|
# * [PyTorch Fullgraph Smoke Test]: Run smoke tests under fullgraph directory, except `test_full_graph.py` as it is a heavy #
|
|
# test that is covered in other steps. Use `find` to launch multiple instances of pytest #
|
|
# so that they do not suffer from: https://github.com/vllm-project/vllm/issues/28965 #
|
|
# * [PyTorch Fullgraph]: #
|
|
# - Limit to no custom ops to reduce running time. Wrap with quotes to escape yaml and avoid starting `-k` string #
|
|
# with a `-` #
|
|
# - Old E2E tests such as: #
|
|
# ```bash #
|
|
# pytest -v -s compile/distributed/test_fusions_e2e.py -k 'TRITON and not +quant_fp8 and not Llama-4' #
|
|
# ``` #
|
|
# were removed in https://github.com/vllm-project/vllm/pull/33293 in favor of new tests in `fusions_e2e`. We #
|
|
# avoid replicating the new jobs in this file as it's deprecated. #
|
|
# * [Basic Models Tests (Extra Initialization) %N]: Only when vLLM model source is modified - test initialization of a #
|
|
# large subset of supported models (the complement of the small subset in #
|
|
# the above test.) Also run if model initialization test file is modified. #
|
|
# * [Language Models Tests (Extra Standard) %N]: Shard slow subset of standard language models tests. Only run when model #
|
|
# source is modified, or when specified test files are modified. #
|
|
# * [Language Models Tests (Hybrid) %N]: Install fast path packages for testing against transformers (mamba, conv1d). #
|
|
# * [Language Models Test (Extended Generation)]: Install fast path packages for testing against transformers (mamba, conv1d). #
|
|
# * [Multi-Modal Models (Standard) 1-4]: #
|
|
# - Do NOT remove `VLLM_WORKER_MULTIPROC_METHOD=spawn` setting as ROCm requires this for certain models to function. #
|
|
# * [Transformers Nightly Models]: Whisper needs `VLLM_WORKER_MULTIPROC_METHOD=spawn` to avoid deadlock. #
|
|
# * [Plugin Tests (2 GPUs)]: #
|
|
# - {`pytest -v -s plugins_tests/test_oot_registration_online.py`}: It needs a clean process #
|
|
# - {`pytest -v -s plugins_tests/test_oot_registration_offline.py`}: It needs a clean process #
|
|
# - {`pytest -v -s plugins_tests/lora_resolvers`}: Unit tests for in-tree lora resolver plugins #
|
|
# * [LoRA TP (Distributed)]: #
|
|
# - There is some Tensor Parallelism related processing logic in LoRA that requires multi-GPU testing for validation. #
|
|
# - {`pytest -v -s -x lora/test_gptoss_tp.py`}: Disabled for now because MXFP4 backend on non-cuda platform doesn't support #
|
|
# LoRA yet. #
|
|
# * [Distributed Tests (NxGPUs)(HW-TAG)]: Don't test llama model here, it seems hf implementation is buggy. See: #
|
|
# https://github.com/vllm-project/vllm/pull/5689 #
|
|
# * [Distributed Tests (NxGPUs)(HW-TAG)]: Some old E2E tests were removed in https://github.com/vllm-project/vllm/pull/33293 #
|
|
# in favor of new tests in fusions_e2e. We avoid replicating the new jobs in #
|
|
# this file as it's deprecated. #
|
|
# #
|
|
#####################################################################################################################################
|
|
|
|
|
|
steps:
|
|
|
|
#########################################################################################################################################
|
|
# #
|
|
# AMD CPU tests #
|
|
# #
|
|
#########################################################################################################################################
|
|
|
|
- label: ":computer: (CPU) Rust Frontend Cargo Style + Clippy" # TBD
|
|
timeout_in_minutes: 35
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx90anightly, amdmi250]
|
|
dind: false
|
|
agent_pool: mi250_1
|
|
no_gpu: true
|
|
optional: true
|
|
working_dir: "/vllm-workspace"
|
|
source_file_dependencies:
|
|
- rust/
|
|
- rust-toolchain.toml
|
|
- .buildkite/test_areas/rust_frontend_cargo.yaml
|
|
- .buildkite/scripts/run-rust-frontend-cargo-ci.sh
|
|
commands:
|
|
- .buildkite/scripts/run-rust-frontend-cargo-ci.sh style-clippy
|
|
|
|
- label: ":computer: (CPU) Rust Frontend Cargo" # TBD
|
|
timeout_in_minutes: 35
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx90anightly, amdmi250]
|
|
dind: false
|
|
agent_pool: mi250_1
|
|
no_gpu: true
|
|
optional: true
|
|
working_dir: "/vllm-workspace"
|
|
source_file_dependencies:
|
|
- rust/
|
|
- rust-toolchain.toml
|
|
- .buildkite/test_areas/rust_frontend_cargo.yaml
|
|
- .buildkite/scripts/run-rust-frontend-cargo-ci.sh
|
|
commands:
|
|
- .buildkite/scripts/run-rust-frontend-cargo-ci.sh test
|
|
|
|
- label: ":computer: (CPU) Docker Build Metadata" # TBD
|
|
timeout_in_minutes: 35
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx90anightly, amdmi250]
|
|
dind: false
|
|
agent_pool: mi250_1
|
|
no_gpu: true
|
|
optional: true
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- .buildkite/scripts/docker-build-metadata-args.sh
|
|
- .buildkite/scripts/ci-bake-rocm.sh
|
|
- .buildkite/release-pipeline.yaml
|
|
- docker/Dockerfile
|
|
- docker/Dockerfile.cpu
|
|
- docker/Dockerfile.rocm
|
|
- docker/Dockerfile.rocm_base
|
|
- docker/ci-rocm.hcl
|
|
- docker/docker-bake.hcl
|
|
- docker/docker-bake-rocm.hcl
|
|
- tests/tools/test_docker_build_metadata_args.py
|
|
commands:
|
|
- pytest -v -s tools/test_docker_build_metadata_args.py
|
|
|
|
- label: ":amd: (MI250) Torch Stable ABI Audit" # TBD
|
|
timeout_in_minutes: 35
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx942nightly, amdmi300]
|
|
dind: false
|
|
agent_pool: mi300_1
|
|
no_gpu: true
|
|
optional: true
|
|
working_dir: "/vllm-workspace/"
|
|
source_file_dependencies:
|
|
- .buildkite/check-torch-abi.py
|
|
- csrc/
|
|
- cmake/
|
|
- setup.py
|
|
- vllm/platforms/rocm.py
|
|
commands:
|
|
- uv pip install --system torch-abi-audit==0.0.1
|
|
- python3 /vllm-workspace/.buildkite/check-torch-abi.py
|
|
|
|
#########################################################################################################################################
|
|
# #
|
|
# MI250 (gfx90a) tests #
|
|
# #
|
|
#########################################################################################################################################
|
|
|
|
#---------------------------------------------------------- mi250 · compile ----------------------------------------------------------#
|
|
|
|
- label: ":amd: (MI250) PyTorch Fullgraph Smoke" # TBD
|
|
timeout_in_minutes: 65
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx90anightly, amdmi250]
|
|
dind: false
|
|
agent_pool: mi250_1
|
|
optional: true
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- vllm/compilation/
|
|
- vllm/model_executor/
|
|
- vllm/v1/attention/
|
|
- vllm/config/compilation.py
|
|
- csrc/
|
|
- tests/compile
|
|
- vllm/platforms/rocm.py
|
|
commands:
|
|
- "find compile/fullgraph/ -name 'test_*.py' -not -name 'test_full_graph.py' -exec pytest -s -v {} \\\\;"
|
|
|
|
#-------------------------------------------------------- mi250 · distributed --------------------------------------------------------#
|
|
|
|
- label: ":amd: (MI250) Pipeline + Context Parallelism" # TBD
|
|
timeout_in_minutes: 90
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx90anightly, amdmi250]
|
|
dind: false
|
|
agent_pool: mi250_4
|
|
num_gpus: 4
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- vllm/distributed/
|
|
- vllm/engine/
|
|
- vllm/executor/
|
|
- vllm/model_executor/models/
|
|
- vllm/model_executor/layers/
|
|
- vllm/v1/attention/backends/
|
|
- vllm/v1/attention/selector.py
|
|
- tests/distributed/
|
|
- vllm/_aiter_ops.py
|
|
- vllm/platforms/rocm.py
|
|
- tests/v1/distributed/test_pp_dp_v2.py
|
|
optional: true
|
|
commands:
|
|
- pytest -v -s distributed/test_pp_cudagraph.py
|
|
- pytest -v -s distributed/test_pipeline_parallel.py
|
|
- pytest -v -s v1/distributed/test_pp_dp_v2.py
|
|
|
|
#---------------------------------------------------------- mi250 · kernels ----------------------------------------------------------#
|
|
|
|
- label: ":amd: (MI250) Helion Kernels Shard %N"
|
|
timeout_in_minutes: 45
|
|
parallelism: 2
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx90anightly, amdmi250]
|
|
dind: false
|
|
agent_pool: mi250_1
|
|
optional: true
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- vllm/utils/import_utils.py
|
|
- tools/benchmark_helion_kernels.py
|
|
- tests/kernels/helion/
|
|
- vllm/platforms/rocm.py
|
|
commands:
|
|
- pip install helion==1.4.0
|
|
- pytest -v -s kernels/helion/ --shard-id=$$BUILDKITE_PARALLEL_JOB --num-shards=$$BUILDKITE_PARALLEL_JOB_COUNT
|
|
|
|
#----------------------------------------------------- mi250 · models / language -----------------------------------------------------#
|
|
|
|
- label: ":amd: (MI250) Language Models (PPL)" # TBD
|
|
timeout_in_minutes: 65
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx90anightly, amdmi250]
|
|
dind: false
|
|
agent_pool: mi250_1
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- vllm/
|
|
- tests/models/language/generation_ppl_test
|
|
- vllm/model_executor/models/qwen3_5.py
|
|
- vllm/model_executor/models/qwen3_5_mtp.py
|
|
- vllm/transformers_utils/configs/qwen3_5.py
|
|
- vllm/transformers_utils/configs/qwen3_5_moe.py
|
|
- vllm/model_executor/models/qwen2.py
|
|
- vllm/model_executor/models/qwen3.py
|
|
- vllm/model_executor/models/qwen3_next.py
|
|
- vllm/model_executor/models/qwen3_next_mtp.py
|
|
- vllm/model_executor/layers/fla/ops/
|
|
- vllm/_aiter_ops.py
|
|
- vllm/v1/attention/backends/triton_attn.py
|
|
- vllm/v1/attention/backends/rocm_attn.py
|
|
- vllm/v1/attention/backends/rocm_aiter_unified_attn.py
|
|
- vllm/v1/attention/backends/rocm_aiter_fa.py
|
|
- vllm/v1/attention/backends/flex_attention.py
|
|
- vllm/v1/attention/ops/
|
|
- vllm/platforms/rocm.py
|
|
optional: true
|
|
commands:
|
|
- pytest -v -s models/language/generation_ppl_test
|
|
|
|
#---------------------------------------------------- mi250 · models / multimodal ----------------------------------------------------#
|
|
|
|
- label: ":amd: (MI250) Multimodal Models (Standard) 3: llava + qwen2_vl" # TBD
|
|
timeout_in_minutes: 55
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx90anightly, amdmi250]
|
|
dind: false
|
|
agent_pool: mi250_1
|
|
optional: true
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- vllm/
|
|
- tests/models/multimodal
|
|
commands:
|
|
- pytest -v -s models/multimodal/generation/test_common.py -m core_model -k "not qwen2 and not qwen3 and not gemma"
|
|
- pytest -v -s models/multimodal/generation/test_qwen2_vl.py -m core_model
|
|
|
|
#------------------------------------------------------------ mi250 · v1 -------------------------------------------------------------#
|
|
|
|
- label: ":amd: (MI250) Batch Invariance" # TBD
|
|
timeout_in_minutes: 35
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx90anightly, amdmi250]
|
|
dind: false
|
|
agent_pool: mi250_1
|
|
optional: true
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- vllm/v1/attention
|
|
- vllm/model_executor/layers
|
|
- tests/v1/determinism/
|
|
- vllm/_aiter_ops.py
|
|
- vllm/platforms/rocm.py
|
|
commands:
|
|
- export VLLM_WORKER_MULTIPROC_METHOD=spawn
|
|
- pip install pytest-timeout pytest-forked
|
|
- pytest -v -s v1/determinism/test_batch_invariance.py
|
|
- pytest -v -s v1/determinism/test_rms_norm_batch_invariant.py
|
|
|
|
- label: ":amd: (MI250) CUDAGraph" # TBD
|
|
timeout_in_minutes: 55
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx90anightly, amdmi250]
|
|
dind: false
|
|
agent_pool: mi250_1
|
|
optional: true
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- tests/v1/cudagraph
|
|
- vllm/v1/cudagraph_dispatcher.py
|
|
- vllm/config/compilation.py
|
|
- vllm/compilation
|
|
- vllm/v1/worker/encoder_cudagraph.py
|
|
- vllm/v1/worker/encoder_cudagraph_defs.py
|
|
- vllm/platforms/rocm.py
|
|
commands:
|
|
- pytest -v -s v1/cudagraph/test_cudagraph_dispatch.py
|
|
- pytest -v -s v1/cudagraph/test_cudagraph_mode.py
|
|
- pytest -v -s v1/cudagraph/test_breakable_cudagraph.py
|
|
- pytest -v -s v1/cudagraph/test_encoder_cudagraph.py
|
|
|
|
- label: ":amd: (MI250) E2E Core" # TBD
|
|
timeout_in_minutes: 50
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx90anightly, amdmi250]
|
|
dind: false
|
|
agent_pool: mi250_1
|
|
optional: true
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- vllm/v1/
|
|
- tests/v1/e2e/general/
|
|
- vllm/platforms/rocm.py
|
|
commands:
|
|
- pytest -v -s v1/e2e/general --ignore v1/e2e/general/test_async_scheduling.py --ignore v1/e2e/general/test_async_scheduling_accuracy.py --ignore v1/e2e/general/test_kv_sharing_fast_prefill.py -k "not test_mamba_prefix_cache_mrv2"
|
|
|
|
- label: ":amd: (MI250) E2E Scheduling" # TBD
|
|
timeout_in_minutes: 55
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx90anightly, amdmi250]
|
|
dind: false
|
|
agent_pool: mi250_1
|
|
optional: true
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- vllm/v1/
|
|
- tests/v1/e2e/general/
|
|
- vllm/platforms/rocm.py
|
|
commands:
|
|
- pytest -v -s v1/e2e/general/test_async_scheduling.py
|
|
|
|
- label: ":amd: (MI250) E2E Scheduling Accuracy" # TBD
|
|
timeout_in_minutes: 55
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx90anightly, amdmi250]
|
|
dind: false
|
|
agent_pool: mi250_1
|
|
optional: true
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- vllm/v1/
|
|
- tests/v1/e2e/general/
|
|
- vllm/platforms/rocm.py
|
|
commands:
|
|
- pytest -v -s v1/e2e/general/test_async_scheduling_accuracy.py
|
|
|
|
- label: ":amd: (MI250) V1 Engine" # TBD
|
|
timeout_in_minutes: 45
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx90anightly, amdmi250]
|
|
dind: false
|
|
agent_pool: mi250_1
|
|
optional: true
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- vllm/v1/
|
|
- tests/v1/engine/
|
|
- tests/v1/test_tensor_ipc_queue.py
|
|
- tests/config/test_mp_reducer.py
|
|
- vllm/config/
|
|
- vllm/engine/arg_utils.py
|
|
- vllm/transformers_utils/config.py
|
|
- vllm/platforms/rocm.py
|
|
- vllm/v1/engine/
|
|
commands:
|
|
- pytest -v -s v1/engine/test_preprocess_error_handling.py
|
|
- pytest -v -s v1/engine --ignore v1/engine/test_preprocess_error_handling.py
|
|
- pytest -v -s v1/test_tensor_ipc_queue.py
|
|
- pytest -v -s config/test_mp_reducer.py
|
|
|
|
- label: ":amd: (MI250) Spec Decode Draft Model" # TBD
|
|
timeout_in_minutes: 60
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx90anightly, amdmi250]
|
|
dind: false
|
|
agent_pool: mi250_1
|
|
optional: true
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- vllm/v1/spec_decode/
|
|
- vllm/v1/worker/gpu/spec_decode/
|
|
- vllm/model_executor/model_loader/
|
|
- vllm/v1/sample/
|
|
- vllm/model_executor/layers/
|
|
- tests/v1/e2e/spec_decode/
|
|
- vllm/platforms/rocm.py
|
|
commands:
|
|
- pytest -v -s v1/e2e/spec_decode/draft_model/
|
|
|
|
- label: ":amd: (MI250) Spec Decode Speculators + MTP" # TBD
|
|
timeout_in_minutes: 75
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx90anightly, amdmi250]
|
|
dind: false
|
|
agent_pool: mi250_1
|
|
optional: true
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- vllm/v1/spec_decode/
|
|
- vllm/v1/worker/gpu/spec_decode/
|
|
- vllm/model_executor/model_loader/
|
|
- vllm/v1/sample/
|
|
- vllm/model_executor/layers/
|
|
- vllm/transformers_utils/configs/speculators/
|
|
- tests/v1/e2e/spec_decode/
|
|
- vllm/platforms/rocm.py
|
|
- vllm/v1/attention/backends/
|
|
commands:
|
|
- pytest -v -s v1/e2e/spec_decode/speculators/
|
|
- pytest -v -s v1/e2e/spec_decode/mtp/
|
|
|
|
#------------------------------------------------------------- mi250 · misc ------------------------------------------------------------#
|
|
|
|
- label: ":amd: (MI300) Python-only Installation" # TBD
|
|
timeout_in_minutes: 55
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx942nightly, amdmi300]
|
|
dind: false
|
|
agent_pool: mi300_1
|
|
optional: true
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- tests/standalone_tests/python_only_compile.sh
|
|
- setup.py
|
|
- vllm/platforms/rocm.py
|
|
commands:
|
|
- bash standalone_tests/python_only_compile.sh
|
|
|
|
#########################################################################################################################################
|
|
# #
|
|
# MI300 (gfx942) tests #
|
|
# #
|
|
#########################################################################################################################################
|
|
|
|
#----------------------------------------------------- mi300 · basic_correctness -----------------------------------------------------#
|
|
|
|
- label: ":amd: (MI300) Distributed Models Shard %N"
|
|
timeout_in_minutes: 80
|
|
parallelism: 3
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx942nightly, amdmi300]
|
|
dind: false
|
|
agent_pool: mi300_2
|
|
num_gpus: 2
|
|
optional: true
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- vllm/models/hy_v4/
|
|
- vllm/config/
|
|
- vllm/v1/worker/gpu/
|
|
- vllm/v1/attention/ops/
|
|
- vllm/v1/attention/backend.py
|
|
- vllm/model_executor/model_loader/sharded_state_loader.py
|
|
- vllm/model_executor/models/
|
|
- vllm/model_executor/layers/
|
|
- vllm/v1/attention/backends/
|
|
- vllm/v1/attention/selector.py
|
|
- vllm/_aiter_ops.py
|
|
- vllm/platforms/rocm.py
|
|
- tests/basic_correctness/
|
|
- tests/model_executor/model_loader/test_sharded_state_loader.py
|
|
- tests/models/
|
|
- vllm/v1/worker/
|
|
- tests/conftest.py
|
|
- tests/utils.py
|
|
- tests/models/test_vision.py
|
|
commands:
|
|
- case "$$BUILDKITE_PARALLEL_JOB" in 0) TARGET_TEST_SUITE=MI300 pytest basic_correctness/ -v -s -m 'distributed(num_gpus=2)' && HIP_VISIBLE_DEVICES=0,1 pytest -v -s model_executor/model_loader/test_sharded_state_loader.py -m '(not slow_test)' ;; 1) pytest models/transformers/test_backend.py -v -s -m 'distributed(num_gpus=2)' && pytest models/language -v -s -m 'distributed(num_gpus=2)' && VLLM_ROCM_USE_AITER=1 pytest -v -s models/test_hyv4_rocm.py -m 'distributed(num_gpus=2)' ;; 2) pytest models/multimodal -v -s -m 'distributed(num_gpus=2)' --ignore models/multimodal/generation/test_whisper.py && VLLM_WORKER_MULTIPROC_METHOD=spawn pytest models/multimodal/generation/test_whisper.py -v -s -m 'distributed(num_gpus=2)' && pytest -v -s models/test_vision.py -m 'distributed(num_gpus=2)' ;; *) exit 2 ;; esac
|
|
|
|
#---------------------------------------------------------- mi300 · compile ----------------------------------------------------------#
|
|
|
|
- label: ":amd: (MI300) PyTorch Compilation" # TBD
|
|
timeout_in_minutes: 90
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx942nightly, amdmi300]
|
|
dind: false
|
|
agent_pool: mi300_1
|
|
optional: true
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- vllm/compilation/
|
|
- vllm/model_executor/layers/
|
|
- vllm/v1/worker/
|
|
- vllm/v1/attention/
|
|
- vllm/v1/cudagraph_dispatcher.py
|
|
- vllm/config/compilation.py
|
|
- csrc/
|
|
- tests/compile
|
|
- vllm/platforms/rocm.py
|
|
- vllm/__init__.py
|
|
- vllm/_aiter_ops.py
|
|
- vllm/_custom_ops.py
|
|
- vllm/config/
|
|
- vllm/distributed/
|
|
- vllm/engine/
|
|
- vllm/env_override.py
|
|
- vllm/envs.py
|
|
- vllm/forward_context.py
|
|
- vllm/inputs/
|
|
- vllm/ir/
|
|
- vllm/kernels/
|
|
- vllm/logger.py
|
|
- vllm/model_executor/
|
|
- vllm/multimodal/
|
|
- vllm/platforms/
|
|
- vllm/plugins/
|
|
- vllm/sampling_params.py
|
|
- vllm/sequence.py
|
|
- vllm/transformers_utils/
|
|
- vllm/triton_utils/
|
|
- vllm/utils/
|
|
- vllm/v1/
|
|
commands:
|
|
- "find compile/ -maxdepth 1 -name 'test_*.py' -print0 | xargs -0 -n1 -I{} pytest -s -v '{}'"
|
|
- pytest -s -v compile/dynamic_shapes/
|
|
|
|
- label: ":amd: (MI300) Fusion E2E Config Sweep" # TBD
|
|
timeout_in_minutes: 40
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx942nightly, amdmi300]
|
|
dind: false
|
|
agent_pool: mi300_1
|
|
num_gpus: 1
|
|
working_dir: "/vllm-workspace/"
|
|
source_file_dependencies:
|
|
- csrc/quantization/
|
|
- vllm/compilation/
|
|
- vllm/model_executor/layers/layernorm.py
|
|
- vllm/model_executor/layers/activation.py
|
|
- vllm/model_executor/layers/attention/attention.py
|
|
- vllm/model_executor/layers/quantization/input_quant_fp8.py
|
|
- tests/compile/fusions_e2e/
|
|
- vllm/_aiter_ops.py
|
|
- vllm/platforms/rocm.py
|
|
optional: true
|
|
commands:
|
|
- rocm-smi
|
|
- pytest -v -s tests/compile/fusions_e2e/test_tp1_quant.py -k "llama-3"
|
|
|
|
- label: ":amd: (MI300) Fusion E2E Quick" # TBD
|
|
timeout_in_minutes: 30
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx942nightly, amdmi300]
|
|
dind: false
|
|
agent_pool: mi300_1
|
|
num_gpus: 1
|
|
working_dir: "/vllm-workspace/"
|
|
source_file_dependencies:
|
|
- csrc/quantization/
|
|
- vllm/model_executor/
|
|
- vllm/v1/attention/
|
|
- vllm/compilation/
|
|
- tests/compile/fusions_e2e/
|
|
- vllm/_aiter_ops.py
|
|
- vllm/platforms/rocm.py
|
|
optional: true
|
|
commands:
|
|
- rocm-smi
|
|
# Run all models and attn backends but only Inductor partition and native custom ops
|
|
- "pytest -v -s tests/compile/fusions_e2e/test_tp1_quant.py -k 'inductor_partition and not +rms_norm and not +quant_fp8'"
|
|
# Different from CUDA, Qwen requires +rms_norm and +quant_fp8 as rms+quant fusion is only supported on AITER
|
|
- "pytest -v -s tests/compile/fusions_e2e/test_tp1_quant.py -k 'inductor_partition and +rms_norm and +quant_fp8 and qwen3'"
|
|
# DeepSeek-Coder uses the portable static-FP8 MLA fusion path on ROCm.
|
|
- "pytest -v -s tests/compile/fusions_e2e/test_tp1_quant.py -k 'inductor_partition and not +rms_norm and +quant_fp8 and DeepSeek-Coder and TRITON_MLA and not True'"
|
|
|
|
- label: ":amd: (MI300) PyTorch Compilation Passes" # TBD
|
|
timeout_in_minutes: 80
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx942nightly, amdmi300]
|
|
dind: false
|
|
agent_pool: mi300_1
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- vllm/
|
|
- tests/compile/passes
|
|
- vllm/platforms/rocm.py
|
|
- vllm/__init__.py
|
|
- vllm/_aiter_ops.py
|
|
- vllm/_custom_ops.py
|
|
- vllm/compilation/
|
|
- vllm/config/
|
|
- vllm/distributed/
|
|
- vllm/engine/
|
|
- vllm/env_override.py
|
|
- vllm/envs.py
|
|
- vllm/forward_context.py
|
|
- vllm/inputs/
|
|
- vllm/ir/
|
|
- vllm/kernels/
|
|
- vllm/logger.py
|
|
- vllm/model_executor/
|
|
- vllm/multimodal/
|
|
- vllm/platforms/
|
|
- vllm/plugins/
|
|
- vllm/sampling_params.py
|
|
- vllm/sequence.py
|
|
- vllm/transformers_utils/
|
|
- vllm/triton_utils/
|
|
- vllm/utils/
|
|
- vllm/v1/
|
|
optional: true
|
|
commands:
|
|
- pytest -s -v compile/passes --ignore compile/passes/distributed
|
|
|
|
- label: ":amd: (MI300) PyTorch Fullgraph" # TBD
|
|
timeout_in_minutes: 90
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx942nightly, amdmi300]
|
|
dind: false
|
|
agent_pool: mi300_1
|
|
optional: true
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- vllm/compilation/
|
|
- vllm/model_executor/
|
|
- vllm/v1/attention/
|
|
- vllm/config/compilation.py
|
|
- csrc/
|
|
- tests/compile
|
|
- vllm/platforms/rocm.py
|
|
- vllm/__init__.py
|
|
- vllm/_aiter_ops.py
|
|
- vllm/_custom_ops.py
|
|
- vllm/config/
|
|
- vllm/distributed/
|
|
- vllm/engine/
|
|
- vllm/env_override.py
|
|
- vllm/envs.py
|
|
- vllm/forward_context.py
|
|
- vllm/inputs/
|
|
- vllm/ir/
|
|
- vllm/kernels/
|
|
- vllm/logger.py
|
|
- vllm/multimodal/
|
|
- vllm/platforms/
|
|
- vllm/plugins/
|
|
- vllm/sampling_params.py
|
|
- vllm/sequence.py
|
|
- vllm/transformers_utils/
|
|
- vllm/triton_utils/
|
|
- vllm/utils/
|
|
- vllm/v1/
|
|
commands:
|
|
- find compile/fullgraph/ -name 'test_*.py' -not -name 'test_full_cudagraph.py' -print0 | xargs -0 -n1 -I{} pytest -s -v '{}'
|
|
|
|
- label: ":amd: (MI300) PyTorch Nightly Dependency Override Check" # TBD
|
|
timeout_in_minutes: 30
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx942nightly, amdmi300]
|
|
dind: false
|
|
agent_pool: mi300_1
|
|
optional: true
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- requirements/test/nightly-torch.txt
|
|
- vllm/platforms/rocm.py
|
|
commands:
|
|
- bash standalone_tests/pytorch_nightly_dependency.sh
|
|
|
|
- label: ":amd: (MI300) Distributed Compile (Inductor Partition)" # TBD
|
|
timeout_in_minutes: 75
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx942nightly, amdmi300]
|
|
dind: false
|
|
agent_pool: mi300_2
|
|
num_gpus: 2
|
|
working_dir: "/vllm-workspace/"
|
|
source_file_dependencies:
|
|
- vllm/compilation/
|
|
- vllm/model_executor/layers
|
|
- tests/compile/passes/distributed/
|
|
- tests/compile/fusions_e2e/
|
|
- vllm/_aiter_ops.py
|
|
- vllm/platforms/rocm.py
|
|
commands:
|
|
- export VLLM_TEST_CLEAN_GPU_MEMORY=1
|
|
- pytest -s -v tests/compile/passes/distributed
|
|
- pytest -v -s tests/compile/fusions_e2e/test_tp2_ar_rms.py::test_tp2_ar_rms_fusions -k "inductor_partition"
|
|
|
|
- label: ":amd: (MI300) Distributed Compile (Dynamo Partition)" # TBD
|
|
timeout_in_minutes: 65
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx942nightly, amdmi300]
|
|
dind: false
|
|
agent_pool: mi300_2
|
|
num_gpus: 2
|
|
working_dir: "/vllm-workspace/"
|
|
source_file_dependencies:
|
|
- vllm/compilation/
|
|
- vllm/model_executor/layers
|
|
- tests/compile/passes/distributed/
|
|
- tests/compile/fusions_e2e/
|
|
- vllm/_aiter_ops.py
|
|
- vllm/platforms/rocm.py
|
|
commands:
|
|
- export VLLM_TEST_CLEAN_GPU_MEMORY=1
|
|
- VLLM_TEST_CLEAN_GPU_MEMORY=1 pytest -v -s tests/compile/passes/distributed/test_async_tp.py
|
|
- pytest -v -s tests/compile/fusions_e2e/test_tp2_ar_rms.py::test_tp2_ar_rms_fusions -k "dynamo_partition"
|
|
|
|
# - label: ":amd: (MI300) Sequence Parallel Correctness" # TBD
|
|
# timeout_in_minutes: 180
|
|
# mirror_hardwares: [amdexperimental, amdproduction, amdgfx942nightly, amdmi300]
|
|
# dind: false
|
|
# agent_pool: mi300_2
|
|
# num_gpus: 2
|
|
# optional: true
|
|
# working_dir: "/vllm-workspace/"
|
|
# source_file_dependencies:
|
|
# - vllm/model_executor/layers/
|
|
# - vllm/compilation/
|
|
# - vllm/v1/worker/
|
|
# - vllm/v1/cudagraph_dispatcher.py
|
|
# - tests/compile/correctness_e2e/test_sequence_parallel.py
|
|
# - vllm/platforms/rocm.py
|
|
# commands:
|
|
# - export VLLM_TEST_CLEAN_GPU_MEMORY=1
|
|
# - pytest -v -s tests/compile/correctness_e2e/test_sequence_parallel.py
|
|
|
|
- label: ":amd: (MI300) Distributed Compile + RPC" # TBD
|
|
timeout_in_minutes: 65
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx942nightly, amdmi300]
|
|
dind: false
|
|
agent_pool: mi300_2
|
|
num_gpus: 2
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- vllm/compilation/
|
|
- vllm/distributed/
|
|
- vllm/engine/
|
|
- vllm/executor/
|
|
- vllm/worker/worker_base.py
|
|
- vllm/v1/engine/
|
|
- vllm/v1/worker/
|
|
- tests/compile/fullgraph/test_basic_correctness.py
|
|
- tests/compile/test_wrapper.py
|
|
- tests/entrypoints/llm/test_collective_rpc.py
|
|
- vllm/platforms/rocm.py
|
|
optional: true
|
|
commands:
|
|
- pytest -v -s entrypoints/llm/test_collective_rpc.py
|
|
- pytest -v -s ./compile/test_wrapper.py
|
|
|
|
#----------------------------------------------------------- mi300 · cuda ------------------------------------------------------------#
|
|
|
|
- label: ":amd: (MI300) CUDA Platform" # TBD
|
|
timeout_in_minutes: 25
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx942nightly, amdmi300]
|
|
dind: false
|
|
agent_pool: mi300_1
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- vllm/
|
|
- tests/cuda
|
|
- vllm/envs.py
|
|
- vllm/logger.py
|
|
- vllm/platforms/
|
|
- vllm/plugins/
|
|
- vllm/utils/
|
|
optional: true
|
|
commands:
|
|
- pytest -v -s cuda/test_cuda_context.py
|
|
- pytest -v -s cuda/test_platform_no_cuda_init.py
|
|
- pytest -v -s cuda/test_cuda_compatibility_path.py
|
|
|
|
#-------------------------------------------------------- mi300 · detokenizer --------------------------------------------------------#
|
|
|
|
- label: ":amd: (MI300) Async Engine, Inputs, Utils, Worker" # TBD
|
|
timeout_in_minutes: 50
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx942nightly, amdmi300]
|
|
dind: false
|
|
agent_pool: mi300_1
|
|
optional: true
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- vllm/
|
|
- tests/detokenizer
|
|
- tests/multimodal
|
|
- tests/utils_
|
|
- vllm/assets/
|
|
- vllm/config/
|
|
- vllm/distributed/
|
|
- vllm/engine/
|
|
- vllm/inputs/
|
|
- vllm/model_executor/
|
|
- vllm/multimodal/
|
|
- vllm/platforms/
|
|
- vllm/sampling_params.py
|
|
- vllm/tokenizers/
|
|
- vllm/transformers_utils/
|
|
- vllm/utils/
|
|
- vllm/v1/
|
|
commands:
|
|
- pytest -v -s detokenizer
|
|
- pytest -v -s -m 'not cpu_test' multimodal
|
|
- pytest -v -s multimodal/test_cache.py::test_sleep_wake_preserves_mm_cache_consistency
|
|
- pytest -v -s utils_
|
|
|
|
#-------------------------------------------------------- mi300 · distributed --------------------------------------------------------#
|
|
|
|
- label: ":amd: (MI300) Distributed Comm Ops" # TBD
|
|
timeout_in_minutes: 30
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx942nightly, amdmi300]
|
|
dind: false
|
|
agent_pool: mi300_2
|
|
num_gpus: 2
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- vllm/distributed
|
|
- tests/distributed
|
|
- vllm/platforms/rocm.py
|
|
commands:
|
|
- pytest -v -s distributed/test_comm_ops.py
|
|
- pytest -v -s distributed/test_shm_broadcast.py
|
|
- pytest -v -s distributed/test_shm_buffer.py
|
|
- pytest -v -s distributed/test_shm_storage.py
|
|
|
|
- label: ":amd: (MI300) EPLB Execution" # TBD
|
|
timeout_in_minutes: 35
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx942nightly, amdmi300]
|
|
dind: false
|
|
agent_pool: mi300_4
|
|
num_gpus: 4
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- vllm/distributed/eplb
|
|
- tests/distributed/test_eplb_execute.py
|
|
- tests/distributed/test_eplb_spec_decode.py
|
|
- vllm/platforms/rocm.py
|
|
commands:
|
|
- pytest -v -s distributed/test_eplb_execute.py
|
|
- pytest -v -s distributed/test_eplb_spec_decode.py
|
|
|
|
- label: ":amd: (MI300) Distributed Features" # TBD
|
|
timeout_in_minutes: 90
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx942nightly, amdmi300]
|
|
dind: false
|
|
agent_pool: mi300_2
|
|
num_gpus: 2
|
|
working_dir: "/vllm-workspace/"
|
|
source_file_dependencies:
|
|
- vllm/distributed/
|
|
- vllm/v1/distributed/
|
|
- vllm/model_executor/layers/fused_moe/
|
|
- vllm/v1/attention/backends/
|
|
- vllm/v1/attention/selector.py
|
|
- tests/v1/distributed/test_dbo.py
|
|
- tests/distributed/test_context_parallel.py
|
|
- examples/features/data_parallel/data_parallel_offline.py
|
|
- vllm/_aiter_ops.py
|
|
- vllm/platforms/rocm.py
|
|
- csrc/custom_quickreduce.cu
|
|
- csrc/ops.h
|
|
- csrc/torch_bindings.cpp
|
|
- vllm/model_executor/layers/
|
|
- vllm/entrypoints/llm.py
|
|
- vllm/config/parallel.py
|
|
- vllm/v1/engine/
|
|
- vllm/v1/executor/
|
|
- vllm/v1/worker/
|
|
- vllm/_custom_ops.py
|
|
- vllm/envs.py
|
|
- tests/distributed/test_rocm_aiter_custom_ar.py
|
|
- tests/distributed/test_rocm_quick_reduce.py
|
|
- tests/distributed/test_quick_all_reduce.py
|
|
- tests/v1/e2e/general/test_rocm_aiter_custom_ar.py
|
|
- tests/utils.py
|
|
- examples/rl/rlhf_async_new_apis.py
|
|
- tests/distributed/test_weight_transfer.py
|
|
- tests/distributed/test_packed_tensor.py
|
|
optional: true
|
|
commands:
|
|
- pytest -v -s tests/distributed/test_context_parallel.py
|
|
- VLLM_ALLOW_INSECURE_SERIALIZATION=1 python3 examples/rl/rlhf_async_new_apis.py
|
|
- VLLM_LOGGING_LEVEL=DEBUG python3 examples/features/data_parallel/data_parallel_offline.py --model=Qwen/Qwen1.5-MoE-A2.7B -tp=1 -dp=2 --max-model-len=2048 --all2all-backend=deepep_high_throughput
|
|
- VLLM_LOGGING_LEVEL=DEBUG python3 examples/features/data_parallel/data_parallel_offline.py --model=Qwen/Qwen1.5-MoE-A2.7B -tp=1 -dp=2 --max-model-len=2048 --all2all-backend=allgather_reducescatter --disable-nccl-for-dp-synchronization
|
|
- pytest -v -s tests/v1/distributed/test_dbo.py
|
|
- VLLM_ALLOW_INSECURE_SERIALIZATION=1 pytest -v -s tests/distributed/test_weight_transfer.py
|
|
- pytest -v -s tests/distributed/test_packed_tensor.py
|
|
- pytest -v -s tests/distributed/test_rocm_aiter_custom_ar.py
|
|
- pytest -v -s tests/v1/e2e/general/test_rocm_aiter_custom_ar.py
|
|
- pytest -v -s tests/distributed/test_rocm_quick_reduce.py
|
|
- pytest -v -s tests/distributed/test_quick_all_reduce.py
|
|
|
|
- label: ":amd: (MI300) Distributed" # TBD
|
|
timeout_in_minutes: 40
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx942nightly, amdmi300]
|
|
dind: false
|
|
agent_pool: mi300_4
|
|
num_gpus: 4
|
|
optional: true
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- vllm/
|
|
- tests/models/test_vision.py
|
|
commands:
|
|
- pytest -v -s distributed/test_custom_all_reduce.py
|
|
- torchrun --nproc_per_node=2 distributed/test_ca_buffer_sharing.py
|
|
- TARGET_TEST_SUITE=MI300 pytest basic_correctness/ -v -s -m 'distributed(num_gpus=2)'
|
|
- pytest -v -s -x lora/test_mixtral.py
|
|
- pytest -v -s models/test_vision.py -m 'distributed(num_gpus=4)'
|
|
|
|
- label: ":amd: (MI300) Distributed Torchrun + Examples" # TBD
|
|
timeout_in_minutes: 55
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx942nightly, amdmi300]
|
|
dind: false
|
|
agent_pool: mi300_4
|
|
num_gpus: 4
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- vllm/distributed/
|
|
- tests/distributed/test_torchrun_example.py
|
|
- tests/distributed/test_torchrun_example_moe.py
|
|
- examples/rl/
|
|
- examples/features/data_parallel/data_parallel_offline.py
|
|
- vllm/platforms/rocm.py
|
|
commands:
|
|
- torchrun --nproc-per-node=4 distributed/test_torchrun_example.py
|
|
- PP_SIZE=2 torchrun --nproc-per-node=4 distributed/test_torchrun_example.py
|
|
- TP_SIZE=4 torchrun --nproc-per-node=4 distributed/test_torchrun_example_moe.py
|
|
- PP_SIZE=2 TP_SIZE=2 torchrun --nproc-per-node=4 distributed/test_torchrun_example_moe.py
|
|
- DP_SIZE=4 ENABLE_EP=1 torchrun --nproc-per-node=4 distributed/test_torchrun_example_moe.py
|
|
- TP_SIZE=2 DP_SIZE=2 ENABLE_EP=1 torchrun --nproc-per-node=4 distributed/test_torchrun_example_moe.py
|
|
- python3 ../examples/features/data_parallel/data_parallel_offline.py --enforce-eager
|
|
# rlhf examples
|
|
- VLLM_ALLOW_INSECURE_SERIALIZATION=1 python3 ../examples/rl/rlhf_http_nccl.py
|
|
- VLLM_ALLOW_INSECURE_SERIALIZATION=1 python3 ../examples/rl/rlhf_http_ipc.py
|
|
|
|
- label: ":amd: (MI300) Elastic EP Scaling" # TBD
|
|
timeout_in_minutes: 55
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx942nightly, amdmi300]
|
|
dind: false
|
|
agent_pool: mi300_4
|
|
num_gpus: 4
|
|
optional: true
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- vllm/distributed/
|
|
- vllm/engine/
|
|
- vllm/executor/
|
|
- vllm/compilation/
|
|
- tests/distributed/
|
|
- vllm/platforms/rocm.py
|
|
commands:
|
|
- uv pip install --system -r /vllm-workspace/requirements/kv_connectors_rocm.txt
|
|
- pytest -v -s distributed/test_elastic_ep.py
|
|
|
|
- label: ":amd: (MI300) RayExecutorV2" # TBD
|
|
timeout_in_minutes: 75
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx942nightly, amdmi300]
|
|
dind: false
|
|
agent_pool: mi300_4
|
|
num_gpus: 4
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- vllm/v1/executor/ray_executor_v2.py
|
|
- vllm/v1/executor/abstract.py
|
|
- vllm/v1/executor/multiproc_executor.py
|
|
- tests/distributed/test_ray_v2_executor.py
|
|
- tests/distributed/test_ray_v2_executor_e2e.py
|
|
- tests/distributed/test_pipeline_parallel.py
|
|
- tests/basic_correctness/models/test_basic_correctness.py
|
|
- vllm/platforms/rocm.py
|
|
commands:
|
|
- export VLLM_USE_RAY_V2_EXECUTOR_BACKEND=1
|
|
- pytest -v -s distributed/test_ray_v2_executor.py
|
|
- pytest -v -s distributed/test_ray_v2_executor_e2e.py
|
|
- pytest -v -s distributed/test_pipeline_parallel.py -k "ray"
|
|
- TARGET_TEST_SUITE=L4 pytest -v -s basic_correctness/models/test_basic_correctness.py -k "ray"
|
|
|
|
- label: ":amd: (MI300) Distributed DP + EP" # TBD
|
|
timeout_in_minutes: 50
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx942nightly, amdmi300]
|
|
dind: false
|
|
agent_pool: mi300_8
|
|
num_gpus: 8
|
|
optional: true
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- examples/features/torchrun/torchrun_dp_example_offline.py
|
|
- vllm/config/parallel.py
|
|
- vllm/distributed/
|
|
- vllm/v1/engine/llm_engine.py
|
|
- vllm/v1/executor/uniproc_executor.py
|
|
- vllm/v1/worker/gpu_worker.py
|
|
- vllm/platforms/rocm.py
|
|
commands:
|
|
- torchrun --nproc-per-node=8 ../examples/features/torchrun/torchrun_dp_example_offline.py --tp-size=2 --pp-size=1 --dp-size=4 --enable-ep
|
|
|
|
- label: ":amd: (MI300) Distributed Torchrun + Shutdown" # TBD
|
|
timeout_in_minutes: 35
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx942nightly, amdmi300]
|
|
dind: false
|
|
agent_pool: mi300_2
|
|
num_gpus: 2
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- vllm/distributed/
|
|
- vllm/engine/
|
|
- vllm/executor/
|
|
- vllm/worker/worker_base.py
|
|
- vllm/v1/engine/
|
|
- vllm/v1/worker/
|
|
- tests/distributed/
|
|
- tests/v1/shutdown
|
|
- tests/v1/worker/test_worker_memory_snapshot.py
|
|
- vllm/platforms/rocm.py
|
|
commands:
|
|
- VLLM_TEST_SAME_HOST=1 torchrun --nproc-per-node=4 distributed/test_same_node.py | grep 'Same node test passed'
|
|
- VLLM_TEST_SAME_HOST=1 VLLM_TEST_WITH_DEFAULT_DEVICE_SET=1 torchrun --nproc-per-node=4 distributed/test_same_node.py | grep 'Same node test passed'
|
|
- HIP_VISIBLE_DEVICES=0,1 pytest -v -s v1/shutdown
|
|
- pytest -v -s v1/worker/test_worker_memory_snapshot.py
|
|
|
|
- label: ":amd: (MI300) Distributed Compile + Comm" # TBD
|
|
timeout_in_minutes: 65
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx942nightly, amdmi300]
|
|
dind: false
|
|
agent_pool: mi300_4
|
|
num_gpus: 4
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- vllm/distributed/
|
|
- tests/distributed/test_pynccl
|
|
- tests/distributed/test_events
|
|
- tests/compile/fullgraph/test_basic_correctness.py
|
|
- tests/distributed/test_symm_mem_allreduce.py
|
|
- tests/distributed/test_multiproc_executor.py
|
|
- vllm/platforms/rocm.py
|
|
commands:
|
|
- pytest -v -s compile/fullgraph/test_basic_correctness.py
|
|
- pytest -v -s distributed/test_pynccl.py
|
|
- pytest -v -s distributed/test_events.py
|
|
- pytest -v -s distributed/test_symm_mem_allreduce.py
|
|
- pytest -v -s distributed/test_multiproc_executor.py::test_multiproc_executor_multi_node
|
|
|
|
#-------------------------------------------------------- mi300 · entrypoints --------------------------------------------------------#
|
|
|
|
- label: ":amd: (MI300) Entrypoints Unit" # TBD
|
|
timeout_in_minutes: 60
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx942nightly, amdmi300]
|
|
dind: false
|
|
agent_pool: mi300_1
|
|
fast_check: true
|
|
optional: true
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- vllm/entrypoints
|
|
- tests/entrypoints/unit_tests
|
|
- tests/entrypoints/weight_transfer
|
|
- tests/entrypoints/launchers
|
|
- vllm/platforms/rocm.py
|
|
commands:
|
|
- pytest -v -s entrypoints/unit_tests
|
|
- pytest -v -s entrypoints/weight_transfer
|
|
- pytest -v -s entrypoints/launchers
|
|
|
|
- label: ":amd: (MI300) Entrypoints Integration (Responses API)" # TBD
|
|
timeout_in_minutes: 50
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx942nightly, amdmi300]
|
|
dind: false
|
|
agent_pool: mi300_1
|
|
optional: true
|
|
fast_check: true
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- vllm/
|
|
- tests/entrypoints/openai/responses
|
|
commands:
|
|
- pytest -v -s entrypoints/openai/responses
|
|
|
|
- label: ":amd: (MI300) Entrypoints Integration (Speech to Text)" # TBD
|
|
timeout_in_minutes: 60
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx942nightly, amdmi300]
|
|
dind: false
|
|
agent_pool: mi300_1
|
|
fast_check: true
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- vllm/
|
|
- tests/entrypoints/speech_to_text
|
|
optional: true
|
|
commands:
|
|
- export VLLM_WORKER_MULTIPROC_METHOD=spawn
|
|
- pytest -v -s entrypoints/speech_to_text
|
|
|
|
- label: ":amd: (MI300) Entrypoints Integration (Multimodal)"
|
|
timeout_in_minutes: 55
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx942nightly, amdmi300]
|
|
dind: false
|
|
agent_pool: mi300_1
|
|
optional: true
|
|
fast_check: true
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- vllm/
|
|
- tests/entrypoints/multimodal
|
|
commands:
|
|
- export VLLM_WORKER_MULTIPROC_METHOD=spawn
|
|
- pytest -v -s entrypoints/multimodal
|
|
|
|
- label: ":amd: (MI300) Scale-out EC Connector E2E"
|
|
timeout_in_minutes: 60
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx942nightly, amdmi300]
|
|
dind: false
|
|
agent_pool: mi300_2
|
|
num_gpus: 2
|
|
optional: true
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- vllm/distributed/ec_transfer/
|
|
- vllm/entrypoints/scale_out/
|
|
- tests/entrypoints/scale_out/ec_integration/
|
|
- vllm/platforms/rocm.py
|
|
commands:
|
|
- export VLLM_WORKER_MULTIPROC_METHOD=spawn
|
|
- bash entrypoints/scale_out/ec_integration/run_scale_out_ec_e2e_test.sh
|
|
|
|
#----------------------------------------------------------- mi300 · evals -----------------------------------------------------------#
|
|
|
|
- label: ":amd: (MI300) LM Eval Small Models" # TBD
|
|
timeout_in_minutes: 45
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx942nightly, amdmi300]
|
|
dind: false
|
|
agent_pool: mi300_1
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- csrc/
|
|
- vllm/model_executor/layers/quantization
|
|
- vllm/model_executor/models/
|
|
- vllm/model_executor/model_loader/
|
|
- vllm/v1/attention/backends/
|
|
- vllm/v1/attention/selector.py
|
|
- vllm/_aiter_ops.py
|
|
- vllm/platforms/rocm.py
|
|
optional: true
|
|
commands:
|
|
- pytest -s -v evals/gsm8k/test_gsm8k_correctness.py --config-list-file=configs/models-small.txt
|
|
|
|
- label: ":amd: (MI300) DeepSeek V2-Lite Prefetch Offload Accuracy" # TBD
|
|
timeout_in_minutes: 30
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx942nightly, amdmi300]
|
|
dind: false
|
|
agent_pool: mi300_1
|
|
num_gpus: 1
|
|
working_dir: "/vllm-workspace"
|
|
source_file_dependencies:
|
|
- vllm/model_executor/models/
|
|
- vllm/model_executor/model_loader/
|
|
- vllm/model_executor/layers/fused_moe/
|
|
- vllm/model_executor/layers/quantization/
|
|
- vllm/v1/attention/backends/
|
|
- vllm/v1/attention/backends/mla/
|
|
- vllm/v1/attention/selector.py
|
|
- .buildkite/scripts/scheduled_integration_test/
|
|
- vllm/_aiter_ops.py
|
|
- vllm/platforms/rocm.py
|
|
optional: true
|
|
commands:
|
|
- bash .buildkite/scripts/scheduled_integration_test/deepseek_v2_lite_prefetch_offload.sh 0.25 200 8030
|
|
|
|
- label: ":amd: (MI300) MRCR Eval Small Models" # TBD
|
|
timeout_in_minutes: 40
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx942nightly, amdmi300]
|
|
dind: false
|
|
agent_pool: mi300_1
|
|
optional: true
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- vllm/model_executor/models/
|
|
- vllm/model_executor/model_loader/
|
|
- vllm/v1/attention/backends/
|
|
- vllm/v1/attention/selector.py
|
|
- vllm/_aiter_ops.py
|
|
- vllm/platforms/rocm.py
|
|
- tests/evals/mrcr/
|
|
commands:
|
|
- pytest -s -v evals/mrcr/test_mrcr_correctness.py --config-list-file=evals/mrcr/configs/models-small.txt
|
|
|
|
- label: ":amd: (MI300) LM Eval KV-Offload" # TBD
|
|
timeout_in_minutes: 55
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx942nightly, amdmi300]
|
|
dind: false
|
|
agent_pool: mi300_1
|
|
optional: true
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- vllm/distributed/kv_transfer/kv_connector/v1/offloading/
|
|
- vllm/distributed/kv_transfer/kv_connector/v1/simple_cpu_offload_connector.py
|
|
- vllm/v1/kv_offload/
|
|
- vllm/v1/simple_kv_offload/
|
|
- tests/evals/gsm8k/test_gsm8k_offloading.py
|
|
- vllm/platforms/rocm.py
|
|
commands:
|
|
- pytest -s -v evals/gsm8k/test_gsm8k_offloading.py -k "nemotron-h-8b or gemma-4-e4b-it"
|
|
|
|
- label: ":amd: (MI300) LM Eval KV-Offload Medium" # TBD
|
|
timeout_in_minutes: 70
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx942nightly, amdmi300]
|
|
dind: false
|
|
agent_pool: mi300_2
|
|
num_gpus: 2
|
|
optional: true
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- vllm/distributed/kv_transfer/kv_connector/v1/offloading/
|
|
- vllm/distributed/kv_transfer/kv_connector/v1/simple_cpu_offload_connector.py
|
|
- vllm/v1/kv_offload/
|
|
- vllm/v1/simple_kv_offload/
|
|
- tests/evals/gsm8k/test_gsm8k_offloading.py
|
|
- vllm/platforms/rocm.py
|
|
commands:
|
|
- pytest -s -v evals/gsm8k/test_gsm8k_offloading.py -k "qwen3.5-35b or deepseek-v2-lite"
|
|
|
|
- label: ":amd: (MI300) LM Eval KV-Offload Large" # TBD
|
|
timeout_in_minutes: 65
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx942nightly, amdmi300]
|
|
dind: false
|
|
agent_pool: mi300_4
|
|
num_gpus: 4
|
|
optional: true
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- vllm/distributed/kv_transfer/kv_connector/v1/offloading/
|
|
- vllm/distributed/kv_transfer/kv_connector/v1/simple_cpu_offload_connector.py
|
|
- vllm/v1/kv_offload/
|
|
- vllm/v1/simple_kv_offload/
|
|
- tests/evals/gsm8k/test_gsm8k_offloading.py
|
|
- vllm/platforms/rocm.py
|
|
commands:
|
|
- VLLM_ROCM_USE_AITER=1 pytest -s -v evals/gsm8k/test_gsm8k_offloading.py -k "deepseek-v4-flash"
|
|
|
|
- label: ":amd: (MI300) LM Eval Small Models Harness" # TBD
|
|
timeout_in_minutes: 35
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx942nightly, amdmi300]
|
|
dind: false
|
|
agent_pool: mi300_1
|
|
optional: true
|
|
working_dir: "/vllm-workspace/.buildkite/lm-eval-harness"
|
|
source_file_dependencies:
|
|
- csrc/
|
|
- vllm/model_executor/layers/quantization
|
|
- vllm/model_executor/models/
|
|
- vllm/model_executor/model_loader/
|
|
- vllm/v1/attention/backends/
|
|
- vllm/v1/attention/selector.py
|
|
- vllm/_aiter_ops.py
|
|
- vllm/platforms/rocm.py
|
|
- .buildkite/lm-eval-harness/
|
|
commands:
|
|
- pytest -s -v test_lm_eval_correctness.py --config-list-file=configs/models-small-rocm.txt
|
|
|
|
- label: ":amd: (MI300) GPQA Eval (GPT-OSS)" # TBD
|
|
timeout_in_minutes: 40
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx942nightly, amdmi300]
|
|
dind: false
|
|
agent_pool: mi300_2
|
|
num_gpus: 2
|
|
optional: true
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- csrc/
|
|
- vllm/model_executor/layers/quantization
|
|
- vllm/model_executor/models/
|
|
- vllm/model_executor/model_loader/
|
|
- vllm/v1/attention/backends/
|
|
- vllm/v1/attention/selector.py
|
|
- vllm/model_executor/layers/fused_moe/
|
|
- vllm/_aiter_ops.py
|
|
- vllm/platforms/rocm.py
|
|
- tests/evals/gpt_oss/
|
|
commands:
|
|
- pytest -s -v evals/gpt_oss/test_gpqa_correctness.py --config-list-file=configs/models-gfx942.txt
|
|
|
|
- label: ":amd: (MI300) LM Eval Small Models Distributed"
|
|
timeout_in_minutes: 55
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx942nightly, amdmi300]
|
|
dind: false
|
|
agent_pool: mi300_2
|
|
num_gpus: 2
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- csrc/
|
|
- vllm/model_executor/layers/quantization
|
|
- vllm/model_executor/models/
|
|
- vllm/model_executor/model_loader/
|
|
- vllm/v1/attention/backends/
|
|
- vllm/v1/attention/selector.py
|
|
- vllm/_aiter_ops.py
|
|
- vllm/platforms/rocm.py
|
|
- tests/evals/gsm8k/configs/models-mi3xx-fp8-and-mixed.txt
|
|
optional: true
|
|
commands:
|
|
- pytest -s -v evals/gsm8k/test_gsm8k_correctness.py --config-list-file=configs/models-mi3xx-fp8-and-mixed.txt
|
|
|
|
- label: ":amd: (MI300) DeepSeek V2-Lite Sync EPLB Accuracy" # TBD
|
|
timeout_in_minutes: 40
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx942nightly, amdmi300]
|
|
dind: false
|
|
agent_pool: mi300_4
|
|
num_gpus: 4
|
|
optional: true
|
|
working_dir: "/vllm-workspace"
|
|
source_file_dependencies:
|
|
- vllm/model_executor/models/
|
|
- vllm/model_executor/model_loader/
|
|
- vllm/distributed/eplb
|
|
- vllm/model_executor/layers/fused_moe/
|
|
- vllm/model_executor/layers/quantization/
|
|
- vllm/v1/attention/backends/
|
|
- vllm/v1/attention/backends/mla/
|
|
- vllm/v1/attention/selector.py
|
|
- .buildkite/scripts/scheduled_integration_test/
|
|
- vllm/_aiter_ops.py
|
|
- vllm/platforms/rocm.py
|
|
commands:
|
|
- VLLM_ENGINE_READY_TIMEOUT_S=1800 bash .buildkite/scripts/scheduled_integration_test/deepseek_v2_lite_ep_eplb.sh 0.25 200 8010
|
|
|
|
- label: ":amd: (MI300) LM Eval Large Models Harness" # TBD
|
|
timeout_in_minutes: 85
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx942nightly, amdmi300]
|
|
dind: false
|
|
agent_pool: mi300_4
|
|
num_gpus: 4
|
|
optional: true
|
|
working_dir: "/vllm-workspace/.buildkite/lm-eval-harness"
|
|
source_file_dependencies:
|
|
- csrc/
|
|
- vllm/model_executor/layers/quantization
|
|
- vllm/model_executor/models/
|
|
- vllm/model_executor/model_loader/
|
|
- vllm/v1/attention/backends/
|
|
- vllm/v1/attention/selector.py
|
|
- vllm/_aiter_ops.py
|
|
- vllm/platforms/rocm.py
|
|
commands:
|
|
- export VLLM_WORKER_MULTIPROC_METHOD=spawn
|
|
- pytest -s -v test_lm_eval_correctness.py --config-list-file=configs/models-large.txt --tp-size=4
|
|
|
|
- label: ":amd: (MI300) Qwen3-30B-A3B-FP8-block Sync EPLB Accuracy" # TBD
|
|
timeout_in_minutes: 30
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx942nightly, amdmi300]
|
|
dind: false
|
|
agent_pool: mi300_4
|
|
num_gpus: 4
|
|
optional: true
|
|
working_dir: "/vllm-workspace"
|
|
source_file_dependencies:
|
|
- vllm/model_executor/models/
|
|
- vllm/model_executor/model_loader/
|
|
- vllm/model_executor/layers/quantization/
|
|
- vllm/distributed/eplb
|
|
- vllm/model_executor/layers/fused_moe/
|
|
- vllm/v1/attention/backends/
|
|
- vllm/v1/attention/selector.py
|
|
- .buildkite/scripts/scheduled_integration_test/
|
|
- vllm/_aiter_ops.py
|
|
- vllm/platforms/rocm.py
|
|
commands:
|
|
- bash .buildkite/scripts/scheduled_integration_test/qwen30b_a3b_fp8_block_ep_eplb.sh 0.8 200 8020
|
|
|
|
- label: ":amd: (MI300) Qwen3-30B-A3B-FP8 DP4 Async EPLB Accuracy" # TBD
|
|
timeout_in_minutes: 30
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx942nightly, amdmi300]
|
|
dind: false
|
|
agent_pool: mi300_4
|
|
num_gpus: 4
|
|
optional: true
|
|
working_dir: "/vllm-workspace"
|
|
source_file_dependencies:
|
|
- vllm/model_executor/models/
|
|
- vllm/model_executor/model_loader/
|
|
- vllm/model_executor/layers/quantization/
|
|
- vllm/distributed/eplb
|
|
- vllm/model_executor/layers/fused_moe/
|
|
- vllm/v1/attention/backends/
|
|
- vllm/v1/attention/selector.py
|
|
- .buildkite/scripts/scheduled_integration_test/
|
|
- vllm/_aiter_ops.py
|
|
- vllm/platforms/rocm.py
|
|
commands:
|
|
- bash .buildkite/scripts/scheduled_integration_test/qwen30b_a3b_fp8_dp4_async_eplb.sh 0.8 200 8050
|
|
|
|
# TODO(akaratza): Retire this standalone ROCm coverage after
|
|
# tests/distributed/test_eplb_spec_decode.py is enabled and validated on ROCm.
|
|
# The upstream parity job currently skips that entire module on ROCm.
|
|
- label: ":amd: (MI300) Qwen3-Next-80B-A3B-Instruct MTP Async EPLB Accuracy" # TBD
|
|
timeout_in_minutes: 45
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx942nightly, amdmi300]
|
|
dind: false
|
|
agent_pool: mi300_4
|
|
num_gpus: 4
|
|
optional: true
|
|
working_dir: "/vllm-workspace"
|
|
source_file_dependencies:
|
|
- vllm/model_executor/models/
|
|
- vllm/model_executor/model_loader/
|
|
- vllm/v1/spec_decode/
|
|
- vllm/distributed/eplb
|
|
- vllm/model_executor/layers/fused_moe/
|
|
- vllm/model_executor/layers/quantization/
|
|
- vllm/v1/attention/backends/
|
|
- vllm/v1/attention/selector.py
|
|
- .buildkite/scripts/scheduled_integration_test/
|
|
- vllm/_aiter_ops.py
|
|
- vllm/platforms/rocm.py
|
|
commands:
|
|
- VLLM_ENGINE_READY_TIMEOUT_S=1800 bash .buildkite/scripts/scheduled_integration_test/qwen3_next_mtp_async_eplb.sh 0.8 1319 8040
|
|
|
|
- label: ":amd: (MI300) LM Eval Large Models" # TBD
|
|
timeout_in_minutes: 65
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx942nightly, amdmi300]
|
|
dind: false
|
|
agent_pool: mi300_8
|
|
optional: true
|
|
num_gpus: 8
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- vllm/model_executor/models/
|
|
- vllm/model_executor/model_loader/
|
|
- vllm/model_executor/layers/quantization/
|
|
- vllm/v1/attention/backends/
|
|
- vllm/v1/attention/selector.py
|
|
- vllm/model_executor/layers/layernorm.py
|
|
- csrc/
|
|
- vllm/_aiter_ops.py
|
|
- vllm/platforms/rocm.py
|
|
- tests/evals/
|
|
commands:
|
|
- export VLLM_WORKER_MULTIPROC_METHOD=spawn
|
|
- export PYTORCH_ROCM_ARCH=gfx942
|
|
- pytest -s -v evals/gsm8k/test_gsm8k_correctness.py --config-list-file=configs/models-mi3xx.txt
|
|
|
|
- label: ":amd: (MI300) LM Eval Large Models ROCm Harness" # TBD
|
|
timeout_in_minutes: 40
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx942nightly, amdmi300]
|
|
dind: false
|
|
agent_pool: mi300_8
|
|
optional: true
|
|
num_gpus: 8
|
|
working_dir: "/vllm-workspace/.buildkite/lm-eval-harness"
|
|
source_file_dependencies:
|
|
- vllm/model_executor/models/
|
|
- vllm/model_executor/model_loader/
|
|
- vllm/model_executor/layers/quantization/
|
|
- vllm/v1/attention/backends/
|
|
- vllm/v1/attention/selector.py
|
|
- vllm/model_executor/layers/layernorm.py
|
|
- csrc/
|
|
- vllm/_aiter_ops.py
|
|
- vllm/platforms/rocm.py
|
|
commands:
|
|
- export VLLM_WORKER_MULTIPROC_METHOD=spawn
|
|
- pytest -s -v test_lm_eval_correctness.py --config-list-file=configs/models-large-rocm.txt --tp-size=8 --timeout=5400
|
|
|
|
- label: ":amd: (MI300) LM Eval Large Models FP8" # TBD
|
|
timeout_in_minutes: 40
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx942nightly, amdmi300]
|
|
dind: false
|
|
agent_pool: mi300_4
|
|
num_gpus: 4
|
|
optional: true
|
|
working_dir: "/vllm-workspace/.buildkite/lm-eval-harness"
|
|
source_file_dependencies:
|
|
- csrc/
|
|
- vllm/model_executor/layers/quantization
|
|
- vllm/model_executor/models/
|
|
- vllm/model_executor/model_loader/
|
|
- vllm/v1/attention/backends/
|
|
- vllm/v1/attention/selector.py
|
|
- vllm/_aiter_ops.py
|
|
- vllm/platforms/rocm.py
|
|
commands:
|
|
- export VLLM_USE_DEEP_GEMM=0
|
|
- pytest -s -v test_lm_eval_correctness.py --config-list-file=configs/models-large-rocm-tp4.txt --tp-size=4
|
|
|
|
#---------------------------------------------------------- mi300 · kernels ----------------------------------------------------------#
|
|
|
|
- label: ":amd: (MI300) Core Operation Kernels Shard %N" # TBD
|
|
timeout_in_minutes: 45
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx942nightly, amdmi300]
|
|
dind: false
|
|
agent_pool: mi300_1
|
|
parallelism: 3
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- csrc/
|
|
- tests/kernels/core
|
|
- tests/kernels/test_top_k_per_row.py
|
|
- tests/kernels/test_concat_mla_q.py
|
|
- tests/kernels/test_rocm_fp32_router_gemm.py
|
|
- vllm/model_executor/layers/fused_moe/router/gate_linear.py
|
|
- vllm/model_executor/layers/fused_moe/router/rocm_fp32_router_gemm.py
|
|
- vllm/model_executor/layers/rotary_embedding/
|
|
- vllm/_aiter_ops.py
|
|
- vllm/platforms/rocm.py
|
|
- tests/kernels/test_fused_qk_norm_rope_gate.py
|
|
optional: true
|
|
commands:
|
|
- pytest -v -s kernels/core --ignore=kernels/core/test_minimax_reduce_rms.py kernels/test_fused_qk_norm_rope_gate.py kernels/test_concat_mla_q.py kernels/test_top_k_per_row.py kernels/test_rocm_fp32_router_gemm.py --shard-id=$$BUILDKITE_PARALLEL_JOB --num-shards=$$BUILDKITE_PARALLEL_JOB_COUNT
|
|
|
|
- label: ":amd: (MI300) Quantization Kernels Shard %N"
|
|
timeout_in_minutes: 120
|
|
parallelism: 6
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx942nightly, amdmi300]
|
|
dind: false
|
|
agent_pool: mi300_1
|
|
optional: true
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- csrc/quantization/
|
|
- csrc/rocm/
|
|
- vllm/model_executor/layers/quantization
|
|
- tests/kernels/quantization
|
|
- tests/kernels/quant_utils.py
|
|
- tests/kernels/utils.py
|
|
- vllm/_aiter_ops.py
|
|
- vllm/kernels/aiter_ops.py
|
|
- vllm/_custom_ops.py
|
|
- vllm/envs.py
|
|
- vllm/platforms/rocm.py
|
|
- vllm/model_executor/kernels/
|
|
- vllm/v1/attention/backends/rocm_aiter_fa.py
|
|
- vllm/config/
|
|
- tests/kernels/quantization/test_rocm_skinny_gemms.py
|
|
commands:
|
|
- pytest -v -s kernels/quantization -k "not test_nvfp4_dense_emulation" --shard-id=$$BUILDKITE_PARALLEL_JOB --num-shards=$$BUILDKITE_PARALLEL_JOB_COUNT
|
|
|
|
- label: ":amd: (MI300) MoE Kernels Shard %N" # TBD
|
|
timeout_in_minutes: 55
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx942nightly, amdmi300]
|
|
dind: false
|
|
agent_pool: mi300_1
|
|
optional: true
|
|
parallelism: 5
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- csrc/quantization/cutlass_w8a8/moe/
|
|
- csrc/moe/
|
|
- csrc/rocm/
|
|
- tests/kernels/moe
|
|
- tests/kernels/utils.py
|
|
- vllm/model_executor/layers/fused_moe/
|
|
- vllm/model_executor/layers/quantization/mxfp4.py
|
|
- vllm/model_executor/layers/quantization/utils/quant_utils.py
|
|
- vllm/distributed/device_communicators/
|
|
- vllm/envs.py
|
|
- vllm/config
|
|
- vllm/_aiter_ops.py
|
|
- vllm/platforms/rocm.py
|
|
commands:
|
|
- pytest -v -s kernels/moe --ignore=kernels/moe/test_modular_oai_triton_moe.py --ignore=kernels/moe/test_modular_kernel_combinations.py --shard-id=$$BUILDKITE_PARALLEL_JOB --num-shards=$$BUILDKITE_PARALLEL_JOB_COUNT
|
|
- pytest -v -s kernels/moe/test_modular_oai_triton_moe.py --shard-id=$$BUILDKITE_PARALLEL_JOB --num-shards=$$BUILDKITE_PARALLEL_JOB_COUNT
|
|
- VLLM_ROCM_USE_AITER=1 VLLM_ROCM_USE_AITER_MOE=1 pytest -v -s kernels/moe/test_modular_kernel_combinations.py --shard-id=$$BUILDKITE_PARALLEL_JOB --num-shards=$$BUILDKITE_PARALLEL_JOB_COUNT
|
|
|
|
- label: ":amd: (MI300) vLLM IR" # TBD
|
|
timeout_in_minutes: 50
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx942nightly, amdmi300]
|
|
dind: false
|
|
agent_pool: mi300_1
|
|
optional: true
|
|
working_dir: "/vllm-workspace/"
|
|
source_file_dependencies:
|
|
- vllm/ir
|
|
- vllm/kernels
|
|
- vllm/_aiter_ops.py
|
|
- vllm/platforms/rocm.py
|
|
- tests/ir/
|
|
- tests/kernels/ir/
|
|
commands:
|
|
- pytest -v -s tests/ir
|
|
- pytest -v -s tests/kernels/ir
|
|
|
|
- label: ":amd: (MI300) Attention Kernels Shard %N" # TBD
|
|
timeout_in_minutes: 90
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx942nightly, amdmi300]
|
|
dind: false
|
|
agent_pool: mi300_1
|
|
optional: true
|
|
parallelism: 2
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- csrc/attention/
|
|
- vllm/v1/attention
|
|
- vllm/model_executor/layers/attention
|
|
- tests/kernels/attention
|
|
- vllm/_aiter_ops.py
|
|
- vllm/envs.py
|
|
- vllm/platforms/rocm.py
|
|
- vllm/utils/flashinfer.py
|
|
commands:
|
|
- pytest -v -s kernels/attention --ignore=kernels/attention/test_triton_unified_attention_diffkv.py --ignore=kernels/attention/test_rocm_aiter_mla_decode.py --ignore=kernels/attention/test_rocm_aiter_mla_decode_metadata.py --ignore=kernels/attention/test_rocm_aiter_mla_fp8_support.py --ignore=kernels/attention/test_rocm_aiter_mla_op_registration.py --shard-id=$$BUILDKITE_PARALLEL_JOB --num-shards=$$BUILDKITE_PARALLEL_JOB_COUNT
|
|
|
|
- label: ":amd: (MI300) MiniMax Reduce RMS Kernels" # TBD
|
|
timeout_in_minutes: 45
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx942nightly, amdmi300]
|
|
dind: false
|
|
agent_pool: mi300_2
|
|
num_gpus: 2
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- vllm/model_executor/layers/minimax_rms_norm/
|
|
- vllm/distributed/
|
|
- vllm/_aiter_ops.py
|
|
- vllm/envs.py
|
|
- vllm/platforms/rocm.py
|
|
- tests/kernels/core/test_minimax_reduce_rms.py
|
|
- tests/utils.py
|
|
- csrc/libtorch_stable/minimax_reduce_rms_kernel.cu
|
|
- csrc/libtorch_stable/minimax_reduce_rms_kernel.h
|
|
- csrc/libtorch_stable/ops.h
|
|
- csrc/libtorch_stable/torch_bindings.cpp
|
|
- vllm/model_executor/layers/mamba/linear/minimax_linear_attn.py
|
|
optional: true
|
|
commands:
|
|
- pytest -v -s kernels/core/test_minimax_reduce_rms.py
|
|
|
|
- label: ":amd: (MI300) Kimi K3" # TBD
|
|
timeout_in_minutes: 65
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx942nightly, amdmi300]
|
|
dind: false
|
|
agent_pool: mi300_1
|
|
optional: true
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- vllm/third_party/flash_linear_attention/ops/kda.py
|
|
- vllm/third_party/flash_linear_attention/ops/chunk_delta_h.py
|
|
- vllm/third_party/flash_linear_attention/ops/l2norm.py
|
|
- vllm/models/kimi_k3/nvidia/kda.py
|
|
- vllm/models/kimi_k3/nvidia/kda_metadata.py
|
|
- vllm/models/kimi_k3/nvidia/ops/third_party/kda/
|
|
- vllm/models/kimi_k3/amd/
|
|
- CMakeLists.txt
|
|
- csrc/libtorch_stable/kimi_k3/fused_kda_decode_kernel_rocm.cu
|
|
- csrc/libtorch_stable/ops.h
|
|
- csrc/libtorch_stable/torch_bindings.cpp
|
|
- tests/models/kimi_k3/test_amd_attn_res.py
|
|
- tests/models/kimi_k3/test_amd_kda_decode.py
|
|
- tests/models/kimi_k3/test_kda.py
|
|
- tests/models/kimi_k3/test_kda_metadata.py
|
|
- vllm/_custom_ops.py
|
|
- vllm/platforms/rocm.py
|
|
commands:
|
|
# Run the ROCm AttnRes/KDA decode counterparts plus existing KDA coverage.
|
|
- pytest -v -s models/kimi_k3/test_amd_attn_res.py models/kimi_k3/test_amd_kda_decode.py models/kimi_k3/test_kda.py models/kimi_k3/test_kda_metadata.py
|
|
|
|
- label: ":amd: (MI355) Kimi K3" # TBD
|
|
timeout_in_minutes: 60
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx950nightly, amdmi355]
|
|
dind: false
|
|
agent_pool: mi355_1
|
|
num_gpus: 1
|
|
optional: true
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- vllm/models/kimi_k3/
|
|
- csrc/libtorch_stable/kimi_k3/
|
|
- tests/models/kimi_k3/
|
|
- vllm/platforms/rocm.py
|
|
commands:
|
|
# Mirrors the ":amd: (MI355) Kimi K3" step generated from
|
|
# .buildkite/test_areas/models_basic.yaml. The ROCm KDA/AttnRes kernels are
|
|
# gated on gfx950, so this has to be MI355. The two kernels/ files the NVIDIA
|
|
# step runs are CUDA-only (they import cuda.bindings.driver), so they are
|
|
# deliberately left out here.
|
|
- pytest -v -s models/kimi_k3
|
|
|
|
- label: ":amd: (MI355) Kimi K3 Multi-GPU" # TBD
|
|
timeout_in_minutes: 30
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx950nightly, amdmi355]
|
|
dind: false
|
|
agent_pool: mi355_4
|
|
num_gpus: 4
|
|
optional: true
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- vllm/models/kimi_k3/
|
|
- tests/models/kimi_k3/
|
|
- vllm/platforms/rocm.py
|
|
commands:
|
|
# The single-GPU step above skips every @multi_gpu_test case. `-m "distributed"`
|
|
# selects exactly those: multi_gpu_test() applies pytest.mark.distributed
|
|
# alongside its skipif (tests/utils.py:2080), so this picks up a new
|
|
# multi-GPU test anywhere in the directory without editing this list.
|
|
# Four GPUs run the two tp4 cases in test_amd_latent_moe_runner.py; the tp8
|
|
# case and the 8/16-GPU cases in test_latent_moe_tail.py still skip.
|
|
- pytest -v -s -m "distributed" models/kimi_k3
|
|
|
|
- label: ":amd: (MI300) Mamba Kernels" # TBD
|
|
timeout_in_minutes: 55
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx942nightly, amdmi300]
|
|
dind: false
|
|
agent_pool: mi300_1
|
|
optional: true
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- csrc/mamba/
|
|
- tests/kernels/mamba
|
|
- vllm/model_executor/layers/mamba/gdn
|
|
- vllm/model_executor/layers/mamba/ops
|
|
- vllm/third_party/flash_linear_attention/
|
|
- vllm/platforms/rocm.py
|
|
commands:
|
|
- pytest -v -s kernels/mamba
|
|
|
|
- label: ":amd: (MI300) DeepEP FP8 MoE Kernels" # TBD
|
|
timeout_in_minutes: 55
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx942nightly, amdmi300]
|
|
dind: false
|
|
agent_pool: mi300_2
|
|
num_gpus: 2
|
|
optional: true
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- csrc/moe/
|
|
- csrc/quantization/w8a8/cutlass/moe/
|
|
- vllm/model_executor/layers/fused_moe/
|
|
- tests/kernels/moe/test_deepep_moe.py
|
|
- vllm/_aiter_ops.py
|
|
- vllm/platforms/rocm.py
|
|
- vllm/envs.py
|
|
commands:
|
|
- pytest -v -s kernels/moe/test_deepep_moe.py
|
|
|
|
#----------------------------------------------------------- mi300 · lora ------------------------------------------------------------#
|
|
|
|
- label: ":amd: (MI300) LoRA TP (Distributed) %N"
|
|
timeout_in_minutes: 65
|
|
parallelism: 4
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx942nightly, amdmi300]
|
|
dind: false
|
|
agent_pool: mi300_4
|
|
num_gpus: 4
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- vllm/lora
|
|
- vllm/model_executor/layers/fused_moe/
|
|
- tests/lora
|
|
- vllm/platforms/rocm.py
|
|
optional: true
|
|
commands:
|
|
- export VLLM_WORKER_MULTIPROC_METHOD=spawn
|
|
- case "$$BUILDKITE_PARALLEL_JOB" in 0|1|2|3) ;; *) echo "unexpected BUILDKITE_PARALLEL_JOB=$$BUILDKITE_PARALLEL_JOB" && exit 1;; esac
|
|
- if [ "$$BUILDKITE_PARALLEL_JOB" = "3" ]; then pytest -v -s -x lora/test_chatglm3_tp.py; fi
|
|
- if [ "$$BUILDKITE_PARALLEL_JOB" = "1" ]; then pytest -v -s -x lora/test_llama_tp.py; fi
|
|
- if [ "$$BUILDKITE_PARALLEL_JOB" = "3" ]; then pytest -v -s -x lora/test_qwen3_with_multi_loras.py; fi
|
|
- if [ "$$BUILDKITE_PARALLEL_JOB" = "2" ]; then pytest -v -s -x lora/test_olmoe_tp.py; fi
|
|
- if [ "$$BUILDKITE_PARALLEL_JOB" = "0" ]; then pytest -v -s -x lora/test_gptoss_tp.py; fi
|
|
- if [ "$$BUILDKITE_PARALLEL_JOB" = "1" ]; then pytest -v -s -x lora/test_qwen35_densemodel_lora.py; fi
|
|
- if [ "$$BUILDKITE_PARALLEL_JOB" = "2" ]; then pytest -v -s -x lora/test_gemma4_tp.py; fi
|
|
- if [ "$$BUILDKITE_PARALLEL_JOB" = "0" ]; then pytest -v -s -x lora/test_sequence_classification.py::test_batched_loras_tp; fi
|
|
|
|
#------------------------------------------------------ mi300 · model_executor -------------------------------------------------------#
|
|
|
|
- label: ":amd: (MI300) Model Executor" # TBD
|
|
timeout_in_minutes: 75
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx942nightly, amdmi300]
|
|
dind: false
|
|
agent_pool: mi300_1
|
|
optional: true
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- vllm/engine/arg_utils.py
|
|
- vllm/config/model.py
|
|
- vllm/model_executor
|
|
- tests/model_executor
|
|
- tests/entrypoints/openai/completion/test_tensorizer_entrypoint.py
|
|
- vllm/_aiter_ops.py
|
|
- vllm/platforms/rocm.py
|
|
- vllm/model_executor/warmup
|
|
- tests/model_executor/test_jit_warmup.py
|
|
- tests/model_executor/test_jit_warmup_cutedsl_launcher.py
|
|
- tests/model_executor/test_jit_warmup_triton_launcher.py
|
|
commands:
|
|
- export VLLM_WORKER_MULTIPROC_METHOD=spawn
|
|
- export PYTHONFAULTHANDLER=1
|
|
- pytest -v -s model_executor -m '(not slow_test)' --timeout=900 --timeout-method=thread
|
|
- pytest -v -s entrypoints/openai/completion/test_tensorizer_entrypoint.py --timeout=900 --timeout-method=thread
|
|
|
|
#---------------------------------------------------- mi300 · model_runner_v2 -------------------------------------------------------#
|
|
|
|
- label: ":amd: (MI300) Model Runner V2 Core" # TBD
|
|
timeout_in_minutes: 55
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx942nightly, amdmi300]
|
|
dind: false
|
|
agent_pool: mi300_1
|
|
optional: true
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- vllm/v1/worker/gpu/
|
|
- vllm/v1/worker/gpu_worker.py
|
|
- vllm/v1/core/sched/
|
|
- vllm/v1/attention/
|
|
- tests/v1/engine/test_llm_engine.py
|
|
- tests/v1/e2e/
|
|
- tests/entrypoints/llm/test_struct_output_generate.py
|
|
- vllm/platforms/rocm.py
|
|
commands:
|
|
- set -x
|
|
- export VLLM_USE_V2_MODEL_RUNNER=1
|
|
- pytest -v -s v1/engine/test_llm_engine.py -k "not test_engine_metrics"
|
|
- pytest -v -s v1/e2e/general/test_async_scheduling.py -k "not ngram"
|
|
- pytest -v -s v1/e2e/general/test_context_length.py
|
|
- pytest -v -s v1/e2e/general/test_min_tokens.py
|
|
- pytest -v -s entrypoints/llm/test_struct_output_generate.py -k "xgrammar and not speculative_config6 and not speculative_config7 and not speculative_config8 and not speculative_config0"
|
|
|
|
- label: ":amd: (MI300) Model Runner V2 Examples" # TBD
|
|
timeout_in_minutes: 40
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx942nightly, amdmi300]
|
|
dind: false
|
|
agent_pool: mi300_1
|
|
optional: true
|
|
working_dir: "/vllm-workspace/examples"
|
|
source_file_dependencies:
|
|
- vllm/v1/worker/gpu/
|
|
- vllm/v1/core/sched/
|
|
- vllm/v1/worker/gpu_worker.py
|
|
- examples/basic/offline_inference/
|
|
- examples/generate/multimodal/
|
|
- examples/features/
|
|
- examples/pooling/embed/vision_embedding_offline.py
|
|
- examples/features/tensorize_vllm_model.py
|
|
- examples/deployment/llm_engine_example.py
|
|
- vllm/platforms/rocm.py
|
|
commands:
|
|
- set -x
|
|
- export VLLM_USE_V2_MODEL_RUNNER=1
|
|
- pip install tensorizer
|
|
- python3 basic/offline_inference/chat.py
|
|
- python3 basic/offline_inference/generate.py --model facebook/opt-125m
|
|
- python3 generate/multimodal/audio_language_offline.py --seed 0
|
|
- python3 generate/multimodal/vision_language_offline.py --seed 0
|
|
- python3 generate/multimodal/vision_language_multi_image_offline.py --seed 0
|
|
- python3 generate/multimodal/encoder_decoder_multimodal_offline.py --model-type whisper --seed 0
|
|
- python3 pooling/embed/vision_embedding_offline.py --seed 0
|
|
- python3 features/automatic_prefix_caching/prefix_caching_offline.py
|
|
- python3 deployment/llm_engine_example.py
|
|
- python3 features/tensorize_vllm_model.py --model facebook/opt-125m serialize --serialized-directory /tmp/ --suffix v1 && python3 features/tensorize_vllm_model.py --model facebook/opt-125m deserialize --path-to-tensors /tmp/vllm/facebook/opt-125m/v1/model.tensors
|
|
- python3 features/speculative_decoding/spec_decode_offline.py --test --method eagle --num_spec_tokens 3 --dataset-name hf --dataset-path philschmid/mt-bench --num-prompts 80 --temp 0 --top-p 1.0 --top-k -1 --tp 1 --enable-chunked-prefill --max-model-len 2048
|
|
- python3 features/speculative_decoding/spec_decode_offline.py --test --method eagle3 --num_spec_tokens 3 --dataset-name hf --dataset-path philschmid/mt-bench --num-prompts 80 --temp 0 --top-p 1.0 --top-k -1 --tp 1 --enable-chunked-prefill --max-model-len 1536
|
|
|
|
- label: ":amd: (MI300) Model Runner V2 Distributed" # TBD
|
|
timeout_in_minutes: 45
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx942nightly, amdmi300]
|
|
dind: false
|
|
agent_pool: mi300_2
|
|
num_gpus: 2
|
|
optional: true
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- vllm/v1/worker/gpu/
|
|
- vllm/v1/worker/gpu_worker.py
|
|
- tests/basic_correctness/models/test_basic_correctness.py
|
|
- tests/v1/distributed/test_async_llm_dp.py
|
|
- tests/v1/distributed/test_eagle_dp.py
|
|
- vllm/platforms/rocm.py
|
|
commands:
|
|
- set -x
|
|
- export VLLM_USE_V2_MODEL_RUNNER=1
|
|
- TARGET_TEST_SUITE=MI300 pytest -v -s basic_correctness/models/test_basic_correctness.py -m 'distributed(num_gpus=2)' -k "not ray and not True"
|
|
- TP_SIZE=1 DP_SIZE=2 pytest -v -s v1/distributed/test_async_llm_dp.py -k "not ray"
|
|
- TP_SIZE=1 DP_SIZE=2 pytest -v -s v1/distributed/test_eagle_dp.py
|
|
|
|
- label: ":amd: (MI300) Model Runner V2 Pipeline Parallelism" # TBD
|
|
timeout_in_minutes: 65
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx942nightly, amdmi300]
|
|
dind: false
|
|
agent_pool: mi300_4
|
|
num_gpus: 4
|
|
optional: true
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- vllm/v1/worker/gpu/
|
|
- vllm/v1/worker/gpu_worker.py
|
|
- tests/distributed/test_pipeline_parallel.py
|
|
- tests/distributed/test_pp_cudagraph.py
|
|
- vllm/platforms/rocm.py
|
|
commands:
|
|
- set -x
|
|
- export VLLM_USE_V2_MODEL_RUNNER=1
|
|
- pytest -v -s distributed/test_pipeline_parallel.py -k "not ray"
|
|
- pytest -v -s distributed/test_pp_cudagraph.py -k "not ray"
|
|
|
|
- label: ":amd: (MI300) Model Runner V2 Spec Decode" # TBD
|
|
timeout_in_minutes: 90
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx942nightly, amdmi300]
|
|
dind: false
|
|
agent_pool: mi300_1
|
|
optional: true
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- vllm/v1/worker/gpu/
|
|
- vllm/v1/worker/gpu_worker.py
|
|
- tests/v1/spec_decode/test_max_len.py
|
|
- tests/v1/spec_decode/test_rejection_sampler_utils.py
|
|
- tests/v1/spec_decode/test_synthetic_rejection_sampler_utils.py
|
|
- tests/v1/e2e/spec_decode/
|
|
- vllm/platforms/rocm.py
|
|
commands:
|
|
- set -x
|
|
- export VLLM_USE_V2_MODEL_RUNNER=1
|
|
- pytest -v -s v1/spec_decode/test_max_len.py -k "eagle or mtp"
|
|
- pytest -v -s v1/spec_decode/test_rejection_sampler_utils.py
|
|
- pytest -v -s v1/spec_decode/test_synthetic_rejection_sampler_utils.py
|
|
- pytest -v -s v1/e2e/spec_decode/eagle/
|
|
- pytest -v -s v1/e2e/spec_decode/speculators/
|
|
- pytest -v -s v1/e2e/spec_decode/mtp/
|
|
|
|
#------------------------------------------------------ mi300 · models / basic -------------------------------------------------------#
|
|
|
|
- label: ":amd: (MI300) Basic Models (Initialization)" # TBD
|
|
timeout_in_minutes: 30
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx942nightly, amdmi300]
|
|
dind: false
|
|
agent_pool: mi300_1
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- vllm/
|
|
- tests/models/test_initialization.py
|
|
- tests/models/registry.py
|
|
optional: true
|
|
commands:
|
|
- pytest -v -s models/test_initialization.py::test_can_initialize_small_subset
|
|
|
|
#----------------------------------------------------- mi300 · models / language -----------------------------------------------------#
|
|
|
|
- label: ":amd: (MI300) Language Models (MTEB)" # TBD
|
|
timeout_in_minutes: 70
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx942nightly, amdmi300]
|
|
dind: false
|
|
agent_pool: mi300_1
|
|
optional: true
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- vllm/
|
|
- tests/models/language/pooling_mteb_test
|
|
commands:
|
|
- pytest -v -s models/language/pooling_mteb_test
|
|
|
|
- label: ":amd: (MI300) Language Models (Extended Pooling) Shard %N"
|
|
timeout_in_minutes: 95
|
|
parallelism: 4
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx942nightly, amdmi300]
|
|
dind: false
|
|
agent_pool: mi300_1
|
|
optional: true
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- vllm/
|
|
- tests/models/language/pooling
|
|
commands:
|
|
- pytest -v -s models/language/pooling -m 'not core_model' --num-shards=$$BUILDKITE_PARALLEL_JOB_COUNT --shard-id=$$BUILDKITE_PARALLEL_JOB
|
|
|
|
- label: ":amd: (MI300) Language Models (Extended Generation)" # TBD
|
|
timeout_in_minutes: 105
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx942nightly, amdmi300]
|
|
dind: false
|
|
agent_pool: mi300_1
|
|
optional: true
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- vllm/
|
|
- tests/models/language/generation
|
|
commands:
|
|
- MAMBA_FORCE_BUILD=TRUE uv pip install --system --no-build-isolation 'git+https://github.com/AndreasKaratzas/mamba@fix-rocm-7.0-warp-size-constexpr'
|
|
- CAUSAL_CONV1D_FORCE_BUILD=TRUE 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)'
|
|
|
|
#---------------------------------------------------- mi300 · models / multimodal ----------------------------------------------------#
|
|
|
|
- label: ":amd: (MI300) Multimodal Models (Extended Generation 1)" # TBD
|
|
timeout_in_minutes: 100
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx942nightly, amdmi300]
|
|
dind: false
|
|
agent_pool: mi300_1
|
|
optional: true
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- vllm/
|
|
- tests/models/multimodal/generation
|
|
- tests/models/multimodal/test_mapping.py
|
|
commands:
|
|
- pytest -v -s models/multimodal/generation -m 'not core_model' --ignore models/multimodal/generation/test_common.py
|
|
- pytest -v -s models/multimodal/test_mapping.py
|
|
|
|
- label: ":amd: (MI300) Multimodal Models (Extended Generation 2) Shard %N"
|
|
timeout_in_minutes: 100
|
|
parallelism: 4
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx942nightly, amdmi300]
|
|
dind: false
|
|
agent_pool: mi300_1
|
|
optional: true
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- vllm/
|
|
- tests/models/multimodal/generation
|
|
commands:
|
|
- pytest -v -s models/multimodal/generation/test_common.py -m 'split(group=0) and not core_model' --num-shards=$$BUILDKITE_PARALLEL_JOB_COUNT --shard-id=$$BUILDKITE_PARALLEL_JOB
|
|
|
|
|
|
- label: ":amd: (MI300) Multimodal Models (Extended Generation 3)" # TBD
|
|
timeout_in_minutes: 90
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx942nightly, amdmi300]
|
|
dind: false
|
|
agent_pool: mi300_1
|
|
optional: true
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- vllm/
|
|
- tests/models/multimodal/generation
|
|
commands:
|
|
- pytest -v -s models/multimodal/generation/test_common.py -m 'split(group=1) and not core_model'
|
|
|
|
- label: ":amd: (MI300) Multimodal Models (Standard) 3: llava + qwen2_vl" # TBD
|
|
timeout_in_minutes: 55
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx942nightly, amdmi300]
|
|
dind: false
|
|
agent_pool: mi300_1
|
|
optional: true
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- vllm/
|
|
- tests/models/multimodal/generation
|
|
- tests/models/multimodal/test_mapping.py
|
|
- tests/models/multimodal
|
|
commands:
|
|
- pytest -v -s models/multimodal/generation/test_common.py -m core_model -k "not qwen2 and not qwen3 and not gemma"
|
|
- pytest -v -s models/multimodal/generation/test_qwen2_vl.py -m core_model
|
|
|
|
- label: ":amd: (MI300) Multimodal Processor Shard %N" # TBD
|
|
timeout_in_minutes: 115
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx942nightly, amdmi300]
|
|
dind: false
|
|
agent_pool: mi300_1
|
|
optional: true
|
|
parallelism: 4
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- vllm/
|
|
- tests/models/multimodal
|
|
- tests/models/registry.py
|
|
commands:
|
|
- pytest -v -s models/multimodal/processing/test_tensor_schema.py --num-shards=$$BUILDKITE_PARALLEL_JOB_COUNT --shard-id=$$BUILDKITE_PARALLEL_JOB
|
|
|
|
- label: ":amd: (MI300) Multimodal Models (Extended Pooling)" # TBD
|
|
timeout_in_minutes: 75
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx942nightly, amdmi300]
|
|
dind: false
|
|
agent_pool: mi300_1
|
|
optional: true
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- vllm/
|
|
- tests/models/multimodal/pooling
|
|
commands:
|
|
- pytest -v -s models/multimodal/pooling -m 'not core_model'
|
|
|
|
#----------------------------------------------------- mi300 · models / quantized -----------------------------------------------------#
|
|
|
|
- label: ":amd: (MI300) Quantized Models" # TBD
|
|
timeout_in_minutes: 90
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx942nightly, amdmi300]
|
|
dind: false
|
|
agent_pool: mi300_1
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- vllm/model_executor/layers/quantization
|
|
- vllm/_aiter_ops.py
|
|
- vllm/platforms/rocm.py
|
|
- tests/models/quantization
|
|
- vllm/model_executor/model_loader/
|
|
optional: true
|
|
commands:
|
|
- unset VLLM_USE_V2_MODEL_RUNNER
|
|
- pytest -v -s models/quantization
|
|
|
|
#-------------------------------------------------- mi300 · models / transformers ---------------------------------------------------#
|
|
|
|
# - label: ":amd: (MI300) Transformers Nightly Models (Initialization) Shard %N" # TBD
|
|
# timeout_in_minutes: 65
|
|
# mirror_hardwares: [amdexperimental, amdproduction, amdgfx942nightly, amdmi300]
|
|
# dind: false
|
|
# agent_pool: mi300_1
|
|
# parallelism: 6
|
|
# optional: true
|
|
# working_dir: "/vllm-workspace/"
|
|
# source_file_dependencies:
|
|
# - vllm/model_executor/models/
|
|
# - vllm/model_executor/model_loader/
|
|
# - vllm/multimodal/
|
|
# - vllm/model_executor/layers/
|
|
# - vllm/v1/attention/backends/
|
|
# - vllm/v1/attention/selector.py
|
|
# - vllm/_aiter_ops.py
|
|
# - vllm/platforms/rocm.py
|
|
# - tests/models/
|
|
# commands:
|
|
# - pip install --upgrade git+https://github.com/huggingface/transformers
|
|
# - pytest -v -s tests/models/test_initialization.py --num-shards=$$BUILDKITE_PARALLEL_JOB_COUNT --shard-id=$$BUILDKITE_PARALLEL_JOB
|
|
|
|
# - label: ":amd: (MI300) Transformers Nightly Models (Processing) Shard %N" # TBD
|
|
# timeout_in_minutes: 145
|
|
# mirror_hardwares: [amdexperimental, amdproduction, amdgfx942nightly, amdmi300]
|
|
# dind: false
|
|
# agent_pool: mi300_1
|
|
# parallelism: 8
|
|
# optional: true
|
|
# working_dir: "/vllm-workspace/"
|
|
# source_file_dependencies:
|
|
# - vllm/model_executor/models/
|
|
# - vllm/model_executor/model_loader/
|
|
# - vllm/multimodal/
|
|
# - vllm/model_executor/layers/
|
|
# - vllm/v1/attention/backends/
|
|
# - vllm/v1/attention/selector.py
|
|
# - vllm/_aiter_ops.py
|
|
# - vllm/platforms/rocm.py
|
|
# - tests/models/
|
|
# commands:
|
|
# - pip install --upgrade git+https://github.com/huggingface/transformers
|
|
# - pytest -v -s tests/models/multimodal/processing/ --num-shards=$$BUILDKITE_PARALLEL_JOB_COUNT --shard-id=$$BUILDKITE_PARALLEL_JOB
|
|
|
|
# - label: ":amd: (MI300) Transformers Nightly Models (Single)" # TBD
|
|
# timeout_in_minutes: 50
|
|
# mirror_hardwares: [amdexperimental, amdproduction, amdgfx942nightly, amdmi300]
|
|
# dind: false
|
|
# agent_pool: mi300_1
|
|
# optional: true
|
|
# working_dir: "/vllm-workspace/"
|
|
# source_file_dependencies:
|
|
# - vllm/model_executor/models/
|
|
# - vllm/model_executor/model_loader/
|
|
# - vllm/multimodal/
|
|
# - vllm/model_executor/layers/
|
|
# - vllm/v1/attention/backends/
|
|
# - vllm/v1/attention/selector.py
|
|
# - vllm/_aiter_ops.py
|
|
# - vllm/platforms/rocm.py
|
|
# - tests/models/
|
|
# - examples/
|
|
# commands:
|
|
# - pip install --upgrade git+https://github.com/huggingface/transformers
|
|
# - pytest -v -s tests/models/transformers/test_backend.py
|
|
# - pytest -v -s tests/models/multimodal/test_mapping.py
|
|
# - python3 examples/basic/offline_inference/chat.py
|
|
# - python3 examples/generate/multimodal/vision_language_offline.py --model-type qwen2_5_vl
|
|
# - VLLM_WORKER_MULTIPROC_METHOD=spawn python3 examples/generate/multimodal/audio_language_offline.py --model-type whisper
|
|
|
|
#---------------------------------------------------------- mi300 · plugins ----------------------------------------------------------#
|
|
|
|
- label: ":amd: (MI300) Plugin Integration" # TBD
|
|
timeout_in_minutes: 50
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx942nightly, amdmi300]
|
|
dind: false
|
|
agent_pool: mi300_2
|
|
num_gpus: 2
|
|
optional: true
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- vllm/plugins/
|
|
- tests/plugins/
|
|
- vllm/platforms/rocm.py
|
|
commands:
|
|
# BEGIN: platform plugin and general plugin tests, all the code in-between runs on dummy platform
|
|
# END: platform plugin tests
|
|
# BEGIN: `io_processor` plugins test, all the code in between uses the `prithvi_io_processor` plugin
|
|
# END: `io_processor` plugins test
|
|
# BEGIN: `bge_m3_sparse io_processor` test
|
|
# END: `bge_m3_sparse io_processor` test
|
|
# BEGIN: `colbert_query io_processor` test
|
|
# END: `colbert_query io_processor` test
|
|
# BEGIN: `stat_logger` plugins test
|
|
# END: `stat_logger` plugins test
|
|
# BEGIN: `endpoint` plugins test
|
|
# END: `endpoint` plugins test
|
|
# BEGIN: other tests
|
|
- pip install -e ./plugins/vllm_add_dummy_platform
|
|
- pytest -v -s plugins_tests/test_platform_plugins.py
|
|
- pip uninstall vllm_add_dummy_platform -y
|
|
- pytest -v -s ./plugins_tests/test_io_processor_plugins.py
|
|
- pip install -e ./plugins/prithvi_io_processor_plugin
|
|
- pytest -v -s plugins_tests/test_terratorch_io_processor_plugins.py
|
|
- pip uninstall prithvi_io_processor_plugin -y
|
|
- pip install -e ./plugins/bge_m3_sparse_plugin
|
|
- pytest -v -s plugins_tests/test_bge_m3_sparse_io_processor_plugins.py
|
|
- pip uninstall bge_m3_sparse_plugin -y
|
|
- pip install -e ./plugins/colbert_query_plugin
|
|
- pytest -v -s plugins_tests/test_colbert_query_io_processor_plugins.py
|
|
- pip uninstall colbert_query_plugin -y
|
|
- pip install -e ./plugins/vllm_add_dummy_stat_logger
|
|
- pytest -v -s plugins_tests/test_stats_logger_plugins.py
|
|
- pip uninstall dummy_stat_logger -y
|
|
- pip install -e ./plugins/vllm_add_dummy_endpoint_plugin
|
|
- pytest -v -s plugins_tests/test_endpoint_plugins.py
|
|
- pip uninstall vllm_add_dummy_endpoint_plugin -y
|
|
- pytest -v -s plugins_tests/test_scheduler_plugins.py
|
|
- pip install -e ./plugins/vllm_add_dummy_model
|
|
- pytest -v -s distributed/test_distributed_oot.py
|
|
- pytest -v -s plugins_tests/test_oot_registration_online.py
|
|
- pytest -v -s plugins_tests/test_oot_registration_offline.py
|
|
- pytest -v -s plugins_tests/lora_resolvers
|
|
|
|
- label: ":amd: (MI300) GGUF Plugin" # TBD
|
|
timeout_in_minutes: 45
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx942nightly, amdmi300]
|
|
dind: false
|
|
agent_pool: mi300_1
|
|
optional: true
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- vllm/model_executor/layers/quantization
|
|
- tests/plugins_tests/test_gguf_plugin.py
|
|
- tests/plugins_tests/gguf
|
|
- vllm/platforms/rocm.py
|
|
commands:
|
|
- pip install "vllm-gguf-plugin >= 0.0.2"
|
|
- pytest -v -s plugins_tests/gguf
|
|
|
|
#------------------------------------------------------- mi300 · rust_frontend -------------------------------------------------------#
|
|
|
|
- label: ":amd: (MI300) Rust Frontend OpenAI Coverage" # TBD
|
|
timeout_in_minutes: 35
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx942nightly, amdmi300]
|
|
dind: false
|
|
agent_pool: mi300_1
|
|
optional: true
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- rust/
|
|
- vllm/benchmarks/
|
|
- vllm/entrypoints/openai/
|
|
- vllm/entrypoints/serve/
|
|
- vllm/v1/sample/
|
|
- tests/utils.py
|
|
- tests/benchmarks/test_serve_cli.py
|
|
- tests/entrypoints/openai/chat_completion/test_chat_completion.py
|
|
- tests/entrypoints/openai/chat_completion/test_chat_logit_bias_validation.py
|
|
- tests/entrypoints/launchers/test_shutdown.py
|
|
- tests/entrypoints/openai/test_return_token_ids.py
|
|
- tests/entrypoints/serve/instrumentator/test_uds.py
|
|
- tests/v1/sample/test_logprobs_e2e.py
|
|
- vllm/platforms/rocm.py
|
|
- tests/entrypoints/openai/chat_completion/test_include_reasoning.py
|
|
- tests/entrypoints/openai/chat_completion/test_serving_chat.py
|
|
- tests/entrypoints/openai/completion/test_shutdown.py
|
|
- tests/entrypoints/openai/test_uds.py
|
|
commands:
|
|
- export VLLM_USE_RUST_FRONTEND=1
|
|
- export VLLM_WORKER_MULTIPROC_METHOD=spawn
|
|
- pytest -v -s benchmarks/test_serve_cli.py -k "not insecure and not (test_bench_serve and not test_bench_serve_chat)"
|
|
- pytest -v -s entrypoints/openai/chat_completion/test_chat_completion.py -k "not test_invalid_json_schema and not test_invalid_regex and not test_kv_transfer_prompt_token_ids_round_trip and not test_kv_transfer_prompt_token_ids_streaming"
|
|
- pytest -v -s entrypoints/openai/chat_completion/test_include_reasoning.py
|
|
- pytest -v -s entrypoints/openai/chat_completion/test_chat_logit_bias_validation.py -k "not multiple"
|
|
- pytest -v -s entrypoints/openai/chat_completion/test_serving_chat.py -k "test_chat_per_request_metrics_follow_server_flag or test_streaming_reasoning_usage_counts_across_deltas or test_completion_tokens_details"
|
|
- pytest -v -s entrypoints/launchers/test_shutdown.py -k "not engine_failure and not test_abort_timeout_exits_quickly"
|
|
- pytest -v -s entrypoints/openai/test_return_token_ids.py -k "not test_comparison"
|
|
- pytest -v -s entrypoints/serve/instrumentator/test_uds.py
|
|
- pytest -v -s v1/sample/test_logprobs_e2e.py -k "test_prompt_logprobs_e2e_server"
|
|
|
|
- label: ":amd: (MI300) Rust Frontend Serve-Admin Coverage"
|
|
timeout_in_minutes: 30
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx942nightly, amdmi300]
|
|
dind: false
|
|
agent_pool: mi300_1
|
|
optional: true
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- rust/
|
|
- vllm/entrypoints/openai/
|
|
- vllm/entrypoints/serve/
|
|
- vllm/v1/engine/
|
|
- tests/utils.py
|
|
- tests/entrypoints/serve/dev/rpc/test_collective_rpc.py
|
|
- tests/entrypoints/scale_out/token_in_token_out/test_serving_tokens.py
|
|
- tests/entrypoints/serve/instrumentator/test_basic.py
|
|
- tests/entrypoints/serve/instrumentator/test_metrics.py
|
|
- tests/entrypoints/serve/tokenize/test_tokenization.py
|
|
- vllm/platforms/rocm.py
|
|
commands:
|
|
- export VLLM_USE_RUST_FRONTEND=1
|
|
- export VLLM_WORKER_MULTIPROC_METHOD=spawn
|
|
- PYTHONPATH=/vllm-workspace pytest -v -s entrypoints/serve/dev/rpc/test_collective_rpc.py
|
|
- pytest -v -s entrypoints/serve/instrumentator/test_basic.py -k "not server_load"
|
|
- pytest -v -s entrypoints/scale_out/token_in_token_out/test_serving_tokens.py -k "not stream and not lora and not test_generate_logprobs and not stop_string_workflow"
|
|
- pytest -v -s entrypoints/serve/instrumentator/test_metrics.py -k "text and not show and not run_batch and not test_metrics_counts and not test_metrics_exist"
|
|
- pytest -v -s entrypoints/serve/tokenize/test_tokenization.py -k "not tokenizer_info"
|
|
|
|
- label: ":amd: (MI300) Rust Frontend Core Correctness" # TBD
|
|
timeout_in_minutes: 25
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx942nightly, amdmi300]
|
|
dind: false
|
|
agent_pool: mi300_1
|
|
optional: true
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- rust/
|
|
- vllm/entrypoints/openai/
|
|
- tests/utils.py
|
|
- tests/entrypoints/openai/correctness/test_lmeval.py
|
|
- vllm/platforms/rocm.py
|
|
commands:
|
|
- export VLLM_USE_RUST_FRONTEND=1
|
|
- export VLLM_WORKER_MULTIPROC_METHOD=spawn
|
|
- pytest -s entrypoints/openai/correctness/test_lmeval.py::test_lm_eval_accuracy_v1_engine
|
|
|
|
- label: ":amd: (MI300) Rust Frontend Tool Use" # TBD
|
|
timeout_in_minutes: 30
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx942nightly, amdmi300]
|
|
dind: false
|
|
agent_pool: mi300_1
|
|
optional: true
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- rust/
|
|
- vllm/entrypoints/openai/
|
|
- vllm/tool_parsers/
|
|
- tests/utils.py
|
|
- tests/tool_use/
|
|
- vllm/platforms/rocm.py
|
|
commands:
|
|
- export VLLM_USE_RUST_FRONTEND=1
|
|
- export VLLM_WORKER_MULTIPROC_METHOD=spawn
|
|
- pytest -v -s tool_use --ignore=tool_use/mistral --models llama3.2 -k "not test_response_format_with_tool_choice_required and not test_parallel_tool_calls_false and not test_tool_call_and_choice"
|
|
|
|
- label: ":amd: (MI300) Rust Frontend Distributed" # TBD
|
|
timeout_in_minutes: 45
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx942nightly, amdmi300]
|
|
dind: false
|
|
agent_pool: mi300_4
|
|
num_gpus: 4
|
|
optional: true
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- rust/
|
|
- vllm/distributed/
|
|
- vllm/engine/
|
|
- vllm/executor/
|
|
- vllm/v1/engine/
|
|
- vllm/v1/worker/
|
|
- tests/utils.py
|
|
- tests/v1/distributed/test_dense_dp_world_size.py
|
|
- tests/v1/distributed/test_external_lb_dp.py
|
|
- tests/v1/distributed/test_hybrid_lb_dp.py
|
|
- tests/v1/distributed/test_internal_lb_dp.py
|
|
- vllm/platforms/rocm.py
|
|
commands:
|
|
- export VLLM_USE_RUST_FRONTEND=1
|
|
- export VLLM_WORKER_MULTIPROC_METHOD=spawn
|
|
- TP_SIZE=1 DP_SIZE=4 pytest -v -s v1/distributed/test_dense_dp_world_size.py
|
|
- VLLM_ENGINE_READY_TIMEOUT_S=1800 TP_SIZE=1 DP_SIZE=4 pytest -v -s v1/distributed/test_internal_lb_dp.py -k "not 4 and not server_info"
|
|
- TP_SIZE=1 DP_SIZE=2 pytest -v -s v1/distributed/test_external_lb_dp.py -k "not 4 and not server_info"
|
|
- TP_SIZE=1 DP_SIZE=4 pytest -v -s v1/distributed/test_hybrid_lb_dp.py -k "not 4 and not server_info"
|
|
|
|
#------------------------------------------------------- mi300 · quantization --------------------------------------------------------#
|
|
|
|
- label: ":amd: (MI300) Quantization Shard %N"
|
|
timeout_in_minutes: 115
|
|
parallelism: 4
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx942nightly, amdmi300]
|
|
dind: false
|
|
agent_pool: mi300_1
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- csrc/
|
|
- vllm/model_executor/layers/quantization
|
|
- vllm/_aiter_ops.py
|
|
- vllm/platforms/rocm.py
|
|
- vllm/model_executor/layers/fused_moe/oracle/unquantized.py
|
|
- vllm/model_executor/layers/fused_moe/unquantized_fused_moe_method.py
|
|
- tests/quantization
|
|
- tests/rocm/test_moe_weight_replay.py
|
|
- tests/rocm/test_moe_padding.py
|
|
optional: true
|
|
commands:
|
|
# temporary install here since we need nightly, will move to requirements/test.in
|
|
# after torchao 0.12 release, and pin a working version of torchao nightly here
|
|
# since torchao nightly is only compatible with torch nightly currently
|
|
# https://github.com/pytorch/ao/issues/2919, we'll have to skip new torchao tests for now
|
|
# we can only upgrade after this is resolved
|
|
# TODO(jerryzh168): resolve the above comment
|
|
- unset VLLM_USE_V2_MODEL_RUNNER
|
|
- uv pip install --system torchao==0.18.0
|
|
- uv pip install --system conch-triton-kernels
|
|
- VLLM_TEST_FORCE_LOAD_FORMAT=auto pytest -v -s quantization/ --ignore quantization/test_blackwell_moe.py --ignore quantization/test_rocm_moe.py --shard-id=$$BUILDKITE_PARALLEL_JOB --num-shards=$$BUILDKITE_PARALLEL_JOB_COUNT
|
|
- if [ "$$BUILDKITE_PARALLEL_JOB" = "0" ]; then pytest -v -s rocm/test_moe_weight_replay.py; fi
|
|
- if [ "$$BUILDKITE_PARALLEL_JOB" = "0" ]; then VLLM_ROCM_USE_AITER=1 VLLM_ROCM_USE_AITER_MOE=1 pytest -v -s rocm/test_moe_padding.py; fi
|
|
|
|
- label: ":amd: (MI300) Quantized Fusions" # TBD
|
|
timeout_in_minutes: 45
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx942nightly, amdmi300]
|
|
dind: false
|
|
agent_pool: mi300_1
|
|
optional: true
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- tests/fusion
|
|
- vllm/model_executor/layers/fusion
|
|
- vllm/model_executor/kernels/linear
|
|
- vllm/model_executor/layers/quantization/compressed_tensors
|
|
- vllm/model_executor/layers/quantization/modelopt.py
|
|
- vllm/platforms/rocm.py
|
|
commands:
|
|
- pytest -v -s fusion/
|
|
|
|
#--------------------------------------------------------- mi300 · samplers ----------------------------------------------------------#
|
|
|
|
- label: ":amd: (MI300) Samplers" # TBD
|
|
timeout_in_minutes: 40
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx942nightly, amdmi300]
|
|
dind: false
|
|
agent_pool: mi300_1
|
|
optional: true
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- vllm/model_executor/layers
|
|
- vllm/sampling_metadata.py
|
|
- vllm/v1/sample/
|
|
- vllm/entrypoints/generate/beam_search/
|
|
- tests/samplers
|
|
- tests/conftest.py
|
|
- vllm/_aiter_ops.py
|
|
- vllm/platforms/rocm.py
|
|
- vllm/entrypoints/generate/beam_search
|
|
commands:
|
|
- pytest -v -s samplers -k "not test_beam_search_passes_multimodal_data"
|
|
|
|
#------------------------------------------------------------ mi300 · misc ------------------------------------------------------------#
|
|
|
|
- label: ":amd: (MI300) Regression" # TBD
|
|
timeout_in_minutes: 25
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx942nightly, amdmi300]
|
|
dind: false
|
|
agent_pool: mi300_1
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- vllm/
|
|
- tests/test_regression
|
|
- vllm/config/
|
|
- vllm/distributed/
|
|
- vllm/engine/
|
|
- vllm/inputs/
|
|
- vllm/model_executor/
|
|
- vllm/multimodal/
|
|
- vllm/platforms/
|
|
- vllm/sampling_params.py
|
|
- vllm/transformers_utils/
|
|
- vllm/utils/
|
|
- vllm/v1/
|
|
optional: true
|
|
commands:
|
|
- pip install 'modelscope<1.38'
|
|
- pytest -v -s test_regression.py
|
|
|
|
#--------------------------------------------------------- mi300 · ray_compat ---------------------------------------------------------#
|
|
|
|
- label: ":amd: (MI300) Ray Dependency Compatibility Check" # TBD
|
|
timeout_in_minutes: 25
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx942nightly, amdmi300]
|
|
dind: false
|
|
agent_pool: mi300_1
|
|
working_dir: "/"
|
|
source_file_dependencies:
|
|
- requirements/
|
|
- setup.py
|
|
- vllm/platforms/rocm.py
|
|
- .buildkite/scripts/check-ray-compatibility.sh
|
|
optional: true
|
|
commands:
|
|
- bash /vllm-workspace/.buildkite/scripts/check-ray-compatibility.sh
|
|
|
|
#------------------------------------------------------------ mi300 · v1 -------------------------------------------------------------#
|
|
|
|
- label: ":amd: (MI300) Spec Decode Speculators + MTP" # TBD
|
|
timeout_in_minutes: 75
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx942nightly, amdmi300]
|
|
dind: false
|
|
agent_pool: mi300_1
|
|
optional: true
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- vllm/v1/spec_decode/
|
|
- vllm/v1/worker/gpu/spec_decode/
|
|
- vllm/model_executor/model_loader/
|
|
- vllm/v1/sample/
|
|
- vllm/model_executor/layers/
|
|
- vllm/transformers_utils/configs/speculators/
|
|
- tests/v1/e2e/spec_decode/
|
|
- vllm/platforms/rocm.py
|
|
- vllm/v1/attention/backends/
|
|
commands:
|
|
- pytest -v -s v1/e2e/spec_decode/speculators/
|
|
- pytest -v -s v1/e2e/spec_decode/mtp/
|
|
|
|
- label: ":amd: (MI300) V1 Spec Decode" # TBD
|
|
timeout_in_minutes: 50
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx942nightly, amdmi300]
|
|
dind: false
|
|
agent_pool: mi300_1
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- vllm/
|
|
- tests/v1/spec_decode
|
|
- vllm/config/
|
|
- vllm/distributed/
|
|
- vllm/inputs/
|
|
- vllm/model_executor/
|
|
- vllm/models/qwen4_exp/
|
|
- vllm/platforms/
|
|
- vllm/sampling_params.py
|
|
- vllm/transformers_utils/
|
|
- vllm/utils/
|
|
- vllm/v1/
|
|
optional: true
|
|
commands:
|
|
- export VLLM_WORKER_MULTIPROC_METHOD=spawn
|
|
- pytest -v -s -m 'not slow_test' v1/spec_decode
|
|
|
|
- label: ":amd: (MI300) Acceptance Length (Large Models)" # TBD
|
|
timeout_in_minutes: 45
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx942nightly, amdmi300]
|
|
dind: false
|
|
agent_pool: mi300_1
|
|
optional: true
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- vllm/v1/spec_decode/
|
|
- vllm/model_executor/models/mlp_speculator.py
|
|
- tests/v1/spec_decode/test_acceptance_length.py
|
|
- vllm/platforms/rocm.py
|
|
commands:
|
|
- export VLLM_ALLOW_INSECURE_SERIALIZATION=1
|
|
- pytest -v -s v1/spec_decode/test_acceptance_length.py -m slow_test
|
|
|
|
- label: ":amd: (MI300) E2E Scheduling" # TBD
|
|
timeout_in_minutes: 55
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx942nightly, amdmi300]
|
|
dind: false
|
|
agent_pool: mi300_1
|
|
optional: true
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- vllm/v1/
|
|
- tests/v1/e2e/general/
|
|
- vllm/platforms/rocm.py
|
|
commands:
|
|
- pytest -v -s v1/e2e/general/test_async_scheduling.py
|
|
|
|
- label: ":amd: (MI300) E2E Scheduling Accuracy" # TBD
|
|
timeout_in_minutes: 55
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx942nightly, amdmi300]
|
|
dind: false
|
|
agent_pool: mi300_1
|
|
optional: true
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- vllm/v1/
|
|
- tests/v1/e2e/general/
|
|
- vllm/platforms/rocm.py
|
|
commands:
|
|
- pytest -v -s v1/e2e/general/test_async_scheduling_accuracy.py
|
|
|
|
- label: ":amd: (MI300) V1 Engine" # TBD
|
|
timeout_in_minutes: 45
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx942nightly, amdmi300]
|
|
dind: false
|
|
agent_pool: mi300_1
|
|
optional: true
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- vllm/v1/engine/
|
|
- tests/v1/engine/
|
|
- tests/v1/test_tensor_ipc_queue.py
|
|
- vllm/platforms/rocm.py
|
|
commands:
|
|
- pytest -v -s v1/engine/test_preprocess_error_handling.py
|
|
- pytest -v -s v1/engine --ignore v1/engine/test_preprocess_error_handling.py
|
|
- pytest -v -s v1/test_tensor_ipc_queue.py
|
|
|
|
- label: ":amd: (MI300) Speculators Correctness Nightly"
|
|
timeout_in_minutes: 40
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx942nightly, amdmi300]
|
|
dind: false
|
|
agent_pool: mi300_1
|
|
optional: true
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- vllm/config/speculative.py
|
|
- vllm/engine/arg_utils.py
|
|
- vllm/transformers_utils/config.py
|
|
- vllm/transformers_utils/configs/speculators/
|
|
- vllm/v1/spec_decode/
|
|
- vllm/v1/worker/gpu/spec_decode/
|
|
- vllm/v1/worker/gpu_model_runner.py
|
|
- vllm/v1/sample/
|
|
- vllm/v1/attention/backends/
|
|
- vllm/v1/attention/selector.py
|
|
- vllm/model_executor/model_loader/
|
|
- vllm/model_executor/layers/
|
|
- vllm/model_executor/models/llama_eagle3.py
|
|
- vllm/model_executor/models/qwen3.py
|
|
- vllm/model_executor/models/qwen3_dflash.py
|
|
- vllm/model_executor/models/registry.py
|
|
- vllm/_aiter_ops.py
|
|
- tests/evals/gsm8k/
|
|
- tests/v1/spec_decode/test_speculators_correctness.py
|
|
- vllm/platforms/rocm.py
|
|
- vllm/v1/worker/gpu/spec_decode/dflash/
|
|
commands:
|
|
- export VLLM_ALLOW_INSECURE_SERIALIZATION=1
|
|
- pytest -v -s v1/spec_decode/test_speculators_correctness.py -m slow_test
|
|
|
|
- label: ":amd: (MI300) Extract Hidden States Integration" # TBD
|
|
timeout_in_minutes: 30
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx942nightly, amdmi300]
|
|
dind: false
|
|
agent_pool: mi300_2
|
|
num_gpus: 2
|
|
working_dir: /vllm-workspace
|
|
source_file_dependencies:
|
|
- vllm/config/speculative.py
|
|
- vllm/distributed/kv_transfer/kv_connector/
|
|
- vllm/model_executor/layers/attention/
|
|
- vllm/model_executor/layers/mamba/
|
|
- vllm/model_executor/model_loader/
|
|
- vllm/model_executor/models/extract_hidden_states.py
|
|
- vllm/model_executor/models/llama.py
|
|
- vllm/model_executor/models/qwen3_5.py
|
|
- vllm/model_executor/models/qwen3_next.py
|
|
- vllm/model_executor/models/registry.py
|
|
- vllm/transformers_utils/configs/extract_hidden_states.py
|
|
- vllm/transformers_utils/configs/qwen3_5.py
|
|
- vllm/v1/attention/backends/
|
|
- vllm/v1/attention/selector.py
|
|
- vllm/v1/kv_cache_interface.py
|
|
- vllm/v1/spec_decode/extract_hidden_states.py
|
|
- vllm/v1/worker/gpu_model_runner.py
|
|
- vllm/_aiter_ops.py
|
|
- tests/v1/kv_connector/extract_hidden_states_integration/
|
|
- vllm/platforms/rocm.py
|
|
- vllm/distributed/kv_transfer/kv_connector/v1/example_hidden_states_connector.py
|
|
- tests/v1/kv_connector/extract_hidden_states_integration
|
|
optional: true
|
|
commands:
|
|
- export VLLM_WORKER_MULTIPROC_METHOD=spawn
|
|
- pytest -v -s -m 'distributed' tests/v1/kv_connector/extract_hidden_states_integration
|
|
|
|
- label: ":amd: (MI300) V1 Attention Shard %N" # TBD
|
|
timeout_in_minutes: 125
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx942nightly, amdmi300]
|
|
dind: false
|
|
agent_pool: mi300_1
|
|
parallelism: 2
|
|
optional: true
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- vllm/config/attention.py
|
|
- vllm/model_executor/layers/attention
|
|
- vllm/v1/attention
|
|
- tests/v1/attention
|
|
- vllm/_aiter_ops.py
|
|
- vllm/envs.py
|
|
- vllm/platforms/rocm.py
|
|
commands:
|
|
- pytest -v -s v1/attention --shard-id=$$BUILDKITE_PARALLEL_JOB --num-shards=$$BUILDKITE_PARALLEL_JOB_COUNT
|
|
|
|
- label: ":amd: (MI300) Metrics, Tracing" # TBD
|
|
timeout_in_minutes: 30
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx942nightly, amdmi300]
|
|
dind: false
|
|
agent_pool: mi300_2
|
|
optional: true
|
|
num_gpus: 2
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- vllm/
|
|
- tests/v1/tracing
|
|
- tests/tracing/
|
|
- vllm/config/
|
|
- vllm/distributed/
|
|
- vllm/engine/
|
|
- vllm/inputs/
|
|
- vllm/model_executor/
|
|
- vllm/multimodal/
|
|
- vllm/platforms/
|
|
- vllm/sampling_params.py
|
|
- vllm/tracing/
|
|
- vllm/transformers_utils/
|
|
- vllm/utils/
|
|
- vllm/v1/
|
|
commands:
|
|
- "pip install \
|
|
'opentelemetry-sdk>=1.26.0' \
|
|
'opentelemetry-api>=1.26.0' \
|
|
'opentelemetry-exporter-otlp>=1.26.0' \
|
|
'opentelemetry-semantic-conventions-ai>=0.4.1'"
|
|
- pytest -v -s v1/tracing
|
|
- pytest -v -s tracing
|
|
|
|
- label: ":amd: (MI300) V1 E2E" # TBD
|
|
timeout_in_minutes: 30
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx942nightly, amdmi300]
|
|
dind: false
|
|
agent_pool: mi300_2
|
|
num_gpus: 2
|
|
optional: true
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- vllm/
|
|
- tests/v1/e2e
|
|
- vllm/compilation/
|
|
- vllm/config/
|
|
- vllm/distributed/
|
|
- vllm/engine/
|
|
- vllm/envs.py
|
|
- vllm/forward_context.py
|
|
- vllm/inputs/
|
|
- vllm/logger.py
|
|
- vllm/logging_utils/
|
|
- vllm/model_executor/
|
|
- vllm/multimodal/
|
|
- vllm/platforms/
|
|
- vllm/sampling_params.py
|
|
- vllm/transformers_utils/
|
|
- vllm/triton_utils/
|
|
- vllm/utils/
|
|
- vllm/v1/
|
|
- tests/v1/e2e/spec_decode/
|
|
commands:
|
|
- >-
|
|
pytest -v -s
|
|
v1/e2e/spec_decode/draft_model/test_draft_model.py::test_draft_model_tensor_parallelism
|
|
v1/e2e/spec_decode/draft_model/test_draft_model.py::test_draft_model_engine_args_tensor_parallelism
|
|
|
|
- label: ":amd: (MI300) Mooncake EC TCP E2E"
|
|
timeout_in_minutes: 60
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx942nightly, amdmi300]
|
|
dind: false
|
|
agent_pool: mi300_2
|
|
num_gpus: 2
|
|
optional: true
|
|
working_dir: "/vllm-workspace"
|
|
source_file_dependencies:
|
|
- vllm/distributed/ec_transfer/
|
|
- vllm/config/ec_transfer.py
|
|
- vllm/config/multimodal.py
|
|
- vllm/config/vllm.py
|
|
- vllm/multimodal/
|
|
- vllm/v1/core/encoder_cache_manager.py
|
|
- vllm/v1/core/sched/
|
|
- vllm/v1/engine/
|
|
- vllm/v1/worker/ec_connector_model_runner_mixin.py
|
|
- vllm/v1/worker/gpu_model_runner.py
|
|
- vllm/v1/worker/gpu_worker.py
|
|
- vllm/v1/worker/gpu/ec_connector.py
|
|
- vllm/v1/worker/gpu/model_runner.py
|
|
- vllm/v1/worker/mm_encoder_model_runner.py
|
|
- vllm/v1/worker/encoder_cudagraph.py
|
|
- vllm/v1/worker/encoder_cudagraph_defs.py
|
|
- vllm/v1/attention/ops/vit_attn_wrappers.py
|
|
- vllm/v1/worker/gpu/model_states/
|
|
- vllm/v1/worker/gpu/mm/
|
|
- tests/v1/cudagraph/test_encoder_cudagraph.py
|
|
- tests/models/multimodal/generation/test_vit_cudagraph.py
|
|
- examples/disaggregated/disaggregated_encoder/
|
|
- tests/v1/ec_connector/
|
|
- requirements/kv_connectors.txt
|
|
- vllm/platforms/rocm.py
|
|
- requirements/rocm.txt
|
|
commands:
|
|
- export VLLM_WORKER_MULTIPROC_METHOD=spawn
|
|
- export PYTHON_BIN='python'
|
|
- export PYTHONUNBUFFERED='1'
|
|
- export VLLM_HOST_IP='127.0.0.1'
|
|
- export VLLM_USE_V2_MODEL_RUNNER='1'
|
|
- export E_CUDAGRAPH_MM_ENCODER='1'
|
|
- export E_MM_ENCODER_ATTN_BACKEND='TRITON_ATTN'
|
|
- export MOONCAKE_EC_PROTOCOL='tcp'
|
|
- export USE_MM_PROMPTS='1'
|
|
- export SKIP_BASELINE='0'
|
|
- export CONCURRENCY='3'
|
|
- export REPEAT='2'
|
|
- export LOG_PATH='/tmp/mooncake-ec-e2e'
|
|
- export BASELINE_FILE='/tmp/mooncake-ec-e2e/baseline.json'
|
|
- bash tests/v1/ec_connector/integration/run_epd_mooncake_ec_full_pipeline.sh
|
|
|
|
- label: ":amd: (MI300) NixlConnector PD edge case" # TBD
|
|
timeout_in_minutes: 65
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx942nightly, amdmi300]
|
|
dind: false
|
|
agent_pool: mi300_2
|
|
num_gpus: 2
|
|
optional: true
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- vllm/distributed/kv_transfer/kv_connector/v1/nixl/
|
|
- vllm/v1/core/sched/
|
|
- tests/v1/kv_connector/nixl_integration/
|
|
- vllm/platforms/rocm.py
|
|
commands:
|
|
- uv pip install --system -r /vllm-workspace/requirements/kv_connectors_rocm.txt
|
|
- PREFILL_GPU_ID=0 DECODE_GPU_ID=1 ATTENTION_BACKEND=TRITON_ATTN bash v1/kv_connector/nixl_integration/run_edge_case_test.sh
|
|
|
|
- label: ":amd: (MI300) CrossLayer KV layout Distributed NixlConnector PD accuracy" # TBD
|
|
timeout_in_minutes: 60
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx942nightly, amdmi300]
|
|
dind: false
|
|
agent_pool: mi300_4
|
|
num_gpus: 4
|
|
optional: true
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- vllm/distributed/kv_transfer/kv_connector/v1/nixl/
|
|
- tests/v1/kv_connector/nixl_integration/
|
|
- vllm/platforms/rocm.py
|
|
commands:
|
|
- uv pip install --system -r /vllm-workspace/requirements/kv_connectors_rocm.txt
|
|
- CROSS_LAYERS_BLOCKS=True ATTENTION_BACKEND=TRITON_ATTN bash v1/kv_connector/nixl_integration/config_sweep_accuracy_test.sh
|
|
|
|
- label: ":amd: (MI300) Distributed DP Extended" # TBD
|
|
timeout_in_minutes: 75
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx942nightly, amdmi300]
|
|
dind: false
|
|
agent_pool: mi300_4
|
|
num_gpus: 4
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- vllm/distributed/
|
|
- tests/v1/distributed
|
|
- tests/v1/engine/test_engine_core_client.py
|
|
- tests/distributed/test_utils
|
|
- vllm/platforms/rocm.py
|
|
commands:
|
|
- export NCCL_CUMEM_HOST_ENABLE=0
|
|
- TP_SIZE=2 DP_SIZE=2 pytest -v -s v1/distributed/test_async_llm_dp.py
|
|
- TP_SIZE=2 DP_SIZE=2 pytest -v -s v1/distributed/test_eagle_dp.py
|
|
- TP_SIZE=2 DP_SIZE=2 pytest -v -s v1/distributed/test_external_lb_dp.py
|
|
- TP_SIZE=1 DP_SIZE=4 pytest -v -s v1/distributed/test_dense_dp_world_size.py
|
|
- TP_SIZE=1 DP_SIZE=4 pytest -v -s v1/distributed/test_internal_lb_dp.py
|
|
- TP_SIZE=1 DP_SIZE=4 pytest -v -s v1/distributed/test_hybrid_lb_dp.py
|
|
- pytest -v -s v1/engine/test_engine_core_client.py::test_kv_cache_events_dp
|
|
- pytest -v -s distributed/test_utils.py
|
|
|
|
- label: ":amd: (MI300) Distributed NixlConnector PD accuracy" # TBD
|
|
timeout_in_minutes: 45
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx942nightly, amdmi300]
|
|
dind: false
|
|
agent_pool: mi300_4
|
|
num_gpus: 4
|
|
optional: true
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- vllm/distributed/kv_transfer/kv_connector/v1/nixl/
|
|
- tests/v1/kv_connector/nixl_integration/
|
|
- vllm/platforms/rocm.py
|
|
commands:
|
|
- uv pip install --system -r /vllm-workspace/requirements/kv_connectors_rocm.txt
|
|
- ATTENTION_BACKEND=TRITON_ATTN bash v1/kv_connector/nixl_integration/config_sweep_accuracy_test.sh
|
|
|
|
- label: ":amd: (MI300) Push NixlConnector PP prefill PD accuracy" # TBD
|
|
timeout_in_minutes: 55
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx942nightly, amdmi300]
|
|
dind: false
|
|
agent_pool: mi300_4
|
|
num_gpus: 4
|
|
optional: true
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- vllm/distributed/kv_transfer/kv_connector/v1/nixl/
|
|
- tests/v1/kv_connector/nixl_integration/
|
|
- tests/v1/kv_connector/nixl_push_integration/
|
|
- vllm/platforms/rocm.py
|
|
commands:
|
|
- uv pip install --system -r /vllm-workspace/requirements/kv_connectors_rocm.txt
|
|
- ATTENTION_BACKEND=TRITON_ATTN bash v1/kv_connector/nixl_push_integration/config_sweep_accuracy_test.sh
|
|
|
|
- label: ":amd: (MI300) DP EP Distributed NixlConnector PD accuracy" # TBD
|
|
timeout_in_minutes: 30
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx942nightly, amdmi300]
|
|
dind: false
|
|
agent_pool: mi300_4
|
|
num_gpus: 4
|
|
optional: true
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- vllm/distributed/kv_transfer/kv_connector/v1/nixl/
|
|
- tests/v1/kv_connector/nixl_integration/
|
|
- vllm/platforms/rocm.py
|
|
commands:
|
|
- uv pip install --system -r /vllm-workspace/requirements/kv_connectors_rocm.txt
|
|
- DP_EP=1 ATTENTION_BACKEND=TRITON_ATTN bash v1/kv_connector/nixl_integration/config_sweep_accuracy_test.sh
|
|
|
|
- label: ":amd: (MI300) Hybrid SSM NixlConnector PD accuracy" # TBD
|
|
timeout_in_minutes: 80
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx942nightly, amdmi300]
|
|
dind: false
|
|
agent_pool: mi300_4
|
|
num_gpus: 4
|
|
optional: true
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- vllm/distributed/kv_transfer/kv_connector/v1/nixl/
|
|
- tests/v1/kv_connector/nixl_integration/
|
|
- vllm/platforms/rocm.py
|
|
commands:
|
|
- uv pip install --system -r /vllm-workspace/requirements/kv_connectors_rocm.txt
|
|
- HYBRID_SSM=1 ATTENTION_BACKEND=TRITON_ATTN bash v1/kv_connector/nixl_integration/config_sweep_accuracy_test.sh
|
|
|
|
- label: ":amd: (MI300) Hybrid SSM NixlConnector PD prefix cache" # TBD
|
|
timeout_in_minutes: 25
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx942nightly, amdmi300]
|
|
dind: false
|
|
agent_pool: mi300_2
|
|
num_gpus: 2
|
|
optional: true
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- vllm/distributed/kv_transfer/kv_connector/v1/nixl/
|
|
- vllm/v1/core/sched/
|
|
- vllm/v1/core/kv_cache_coordinator.py
|
|
- tests/v1/kv_connector/nixl_integration/
|
|
- vllm/platforms/rocm.py
|
|
commands:
|
|
- uv pip install --system -r /vllm-workspace/requirements/kv_connectors_rocm.txt
|
|
- ATTENTION_BACKEND=TRITON_ATTN bash v1/kv_connector/nixl_integration/run_mamba_prefix_cache_test.sh
|
|
|
|
- label: ":amd: (MI300) MultiConnector (Nixl+Offloading) PD accuracy" # TBD
|
|
timeout_in_minutes: 30
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx942nightly, amdmi300]
|
|
dind: false
|
|
agent_pool: mi300_2
|
|
num_gpus: 2
|
|
optional: true
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- vllm/distributed/kv_transfer/kv_connector/v1/nixl/
|
|
- vllm/distributed/kv_transfer/kv_connector/v1/multi_connector.py
|
|
- vllm/distributed/kv_transfer/kv_connector/v1/offloading_connector.py
|
|
- vllm/distributed/kv_transfer/kv_connector/v1/offloading/
|
|
- tests/v1/kv_connector/nixl_integration/
|
|
- vllm/platforms/rocm.py
|
|
commands:
|
|
- uv pip install --system -r /vllm-workspace/requirements/kv_connectors_rocm.txt
|
|
- ATTENTION_BACKEND=TRITON_ATTN bash v1/kv_connector/nixl_integration/run_multi_connector_accuracy_test.sh
|
|
|
|
- label: ":amd: (MI300) MultiConnector (Nixl+Offloading) PD edge cases" # TBD
|
|
timeout_in_minutes: 25
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx942nightly, amdmi300]
|
|
dind: false
|
|
agent_pool: mi300_2
|
|
num_gpus: 2
|
|
optional: true
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- vllm/distributed/kv_transfer/kv_connector/v1/nixl/
|
|
- vllm/distributed/kv_transfer/kv_connector/v1/multi_connector.py
|
|
- vllm/distributed/kv_transfer/kv_connector/v1/offloading_connector.py
|
|
- vllm/distributed/kv_transfer/kv_connector/v1/offloading/
|
|
- tests/v1/kv_connector/nixl_integration/
|
|
- vllm/platforms/rocm.py
|
|
commands:
|
|
- uv pip install --system -r /vllm-workspace/requirements/kv_connectors_rocm.txt
|
|
- ATTENTION_BACKEND=TRITON_ATTN bash v1/kv_connector/nixl_integration/run_multi_connector_edge_case_test.sh
|
|
|
|
- label: ":amd: (MI300) V1 E2E Hybrid Chunked Prefill" # TBD
|
|
timeout_in_minutes: 35
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx942nightly, amdmi300]
|
|
dind: false
|
|
agent_pool: mi300_4
|
|
num_gpus: 4
|
|
optional: true
|
|
source_file_dependencies:
|
|
- vllm/v1/attention/backends/utils.py
|
|
- vllm/v1/worker/gpu_model_runner.py
|
|
- tests/v1/e2e/test_hybrid_chunked_prefill.py
|
|
- vllm/v1/core/
|
|
working_dir: /vllm-workspace/tests
|
|
commands:
|
|
- pytest -v -s v1/e2e/test_hybrid_chunked_prefill.py
|
|
|
|
#########################################################################################################################################
|
|
# #
|
|
# MI355 (gfx950) tests #
|
|
# #
|
|
#########################################################################################################################################
|
|
|
|
#----------------------------------------------------- mi355 · basic_correctness -----------------------------------------------------#
|
|
|
|
- label: ":amd: (MI355 DPX) Basic Correctness Models %N" # TBD
|
|
timeout_in_minutes: 60
|
|
parallelism: 2
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx950nightly, amdmi355]
|
|
dind: false
|
|
agent_pool: mi355_dpx
|
|
fast_check: true
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- vllm/
|
|
- tests/basic_correctness/
|
|
optional: true
|
|
commands:
|
|
- export VLLM_WORKER_MULTIPROC_METHOD=spawn
|
|
- VLLM_TARGET_TEST_SUITE=MI355 pytest -v -s basic_correctness/models --num-shards=$$BUILDKITE_PARALLEL_JOB_COUNT --shard-id=$$BUILDKITE_PARALLEL_JOB
|
|
|
|
- label: ":amd: (MI355 DPX) Basic Correctness CuMem" # TBD
|
|
timeout_in_minutes: 60
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx950nightly, amdmi355]
|
|
dind: false
|
|
agent_pool: mi355_dpx
|
|
fast_check: true
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- vllm/
|
|
- tests/basic_correctness/
|
|
optional: true
|
|
commands:
|
|
- export VLLM_WORKER_MULTIPROC_METHOD=spawn
|
|
- pytest -v -s basic_correctness/memory/cumem
|
|
|
|
- label: ":amd: (MI300) Basic Correctness Sleep Mode" # TBD
|
|
timeout_in_minutes: 60
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx942nightly, amdmi300]
|
|
dind: false
|
|
agent_pool: mi300_1
|
|
fast_check: true
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- vllm/
|
|
- tests/basic_correctness/
|
|
optional: true
|
|
commands:
|
|
- export VLLM_WORKER_MULTIPROC_METHOD=spawn
|
|
- pytest -v -s basic_correctness/memory/sleep_mode
|
|
|
|
- label: ":amd: (MI355 DPX) Basic Correctness CPU Offload" # TBD
|
|
timeout_in_minutes: 60
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx950nightly, amdmi355]
|
|
dind: false
|
|
agent_pool: mi355_dpx
|
|
fast_check: true
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- vllm/
|
|
- tests/basic_correctness/
|
|
optional: true
|
|
commands:
|
|
- export VLLM_WORKER_MULTIPROC_METHOD=spawn
|
|
- pytest -v -s basic_correctness/cpu_offload
|
|
|
|
- label: ":amd: (MI355 DPX) Basic Correctness Prefetch Offload" # TBD
|
|
timeout_in_minutes: 60
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx950nightly, amdmi355]
|
|
dind: false
|
|
agent_pool: mi355_dpx
|
|
fast_check: true
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- vllm/
|
|
- tests/basic_correctness/
|
|
optional: true
|
|
commands:
|
|
- export VLLM_WORKER_MULTIPROC_METHOD=spawn
|
|
- pytest -v -s basic_correctness/prefetch_offload
|
|
|
|
#-------------------------------------------------------- mi355 · benchmarks ---------------------------------------------------------#
|
|
|
|
- label: ":amd: (MI355 DPX) Benchmarks CLI" # TBD
|
|
timeout_in_minutes: 40
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx950nightly, amdmi355]
|
|
dind: false
|
|
agent_pool: mi355_dpx
|
|
optional: true
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- vllm/
|
|
- tests/benchmarks/
|
|
- rust/src/bench/tests/python_serve_flags.txt
|
|
commands:
|
|
- pytest -v -s benchmarks/
|
|
|
|
- label: ":amd: (MI355) Attention Benchmark Smoke" # TBD
|
|
timeout_in_minutes: 25
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx950nightly, amdmi355]
|
|
dind: false
|
|
agent_pool: mi355_2
|
|
num_gpus: 2
|
|
optional: true
|
|
working_dir: "/vllm-workspace/"
|
|
source_file_dependencies:
|
|
- benchmarks/attention_benchmarks/
|
|
- vllm/v1/attention/
|
|
- vllm/_aiter_ops.py
|
|
- vllm/platforms/rocm.py
|
|
commands:
|
|
- python3 benchmarks/attention_benchmarks/benchmark.py --backends ROCM_ATTN ROCM_AITER_FA ROCM_AITER_UNIFIED_ATTN --batch-specs "8q1s1k"
|
|
|
|
#---------------------------------------------------------- mi355 · compile ----------------------------------------------------------#
|
|
|
|
- label: ":amd: (MI355 DPX) PyTorch Compilation H100 Cases"
|
|
timeout_in_minutes: 45
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx950nightly, amdmi355]
|
|
dind: false
|
|
agent_pool: mi355_dpx
|
|
num_gpus: 1
|
|
optional: true
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- vllm/__init__.py
|
|
- vllm/_aiter_ops.py
|
|
- vllm/_custom_ops.py
|
|
- vllm/compilation/
|
|
- vllm/config/
|
|
- vllm/distributed/
|
|
- "!vllm/distributed/kv_transfer/"
|
|
- vllm/engine/
|
|
- vllm/env_override.py
|
|
- vllm/envs.py
|
|
- vllm/forward_context.py
|
|
- vllm/inputs/
|
|
- vllm/ir/
|
|
- vllm/kernels/
|
|
- vllm/logger.py
|
|
- vllm/model_executor/
|
|
- vllm/multimodal/
|
|
- vllm/platforms/
|
|
- vllm/plugins/
|
|
- vllm/sampling_params.py
|
|
- vllm/sequence.py
|
|
- vllm/transformers_utils/
|
|
- vllm/triton_utils/
|
|
- vllm/utils/
|
|
- vllm/v1/
|
|
- tests/compile/h100/
|
|
- tests/utils.py
|
|
- csrc/
|
|
commands:
|
|
- pytest -v -s compile/h100/test_startup.py
|
|
|
|
- label: ":amd: (MI355) Fusion E2E TP2 AR-RMS FP8 Config Sweep"
|
|
timeout_in_minutes: 45
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx950nightly, amdmi355]
|
|
dind: false
|
|
agent_pool: mi355_2
|
|
num_gpus: 2
|
|
optional: true
|
|
working_dir: "/vllm-workspace/"
|
|
source_file_dependencies:
|
|
- vllm/compilation/
|
|
- vllm/model_executor/
|
|
- vllm/distributed/
|
|
- vllm/v1/attention/
|
|
- vllm/_aiter_ops.py
|
|
- vllm/platforms/rocm.py
|
|
- tests/compile/fusions_e2e/
|
|
- tests/utils.py
|
|
- tests/conftest.py
|
|
- csrc/quantization/
|
|
commands:
|
|
# Existing MI300 jobs own BF16; TP2 Quick owns the excluded Llama cell.
|
|
- amd-smi
|
|
- VLLM_ROCM_USE_AITER=1 VLLM_ROCM_USE_AITER_CUSTOM_AR=1 pytest -v -s tests/compile/fusions_e2e/test_tp2_ar_rms.py -k 'test_tp2_ar_rms_fp8_fusions and Llama-3 and ROCM_AITER_UNIFIED_ATTN and (dynamo_partition or +quant_fp8 or +rms_norm)'
|
|
|
|
- label: ":amd: (MI355) Fusion E2E TP2 FP8 BF16 rewrites"
|
|
timeout_in_minutes: 45
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx950nightly, amdmi355]
|
|
dind: false
|
|
agent_pool: mi355_2
|
|
num_gpus: 2
|
|
optional: true
|
|
working_dir: "/vllm-workspace/"
|
|
source_file_dependencies:
|
|
- vllm/compilation/
|
|
- vllm/model_executor/
|
|
- vllm/distributed/
|
|
- vllm/v1/attention/
|
|
- vllm/_aiter_ops.py
|
|
- vllm/platforms/rocm.py
|
|
- tests/compile/fusions_e2e/
|
|
- tests/utils.py
|
|
- tests/conftest.py
|
|
- csrc/quantization/
|
|
commands:
|
|
# Dummy-weight rewrite coverage; trained FP8 AR/RMS equivalence is unresolved.
|
|
# TODO: FP4 output-quant and trained MXFP4 AR/RMS fusion remain unsupported.
|
|
- amd-smi
|
|
- VLLM_ROCM_USE_AITER=1 VLLM_ROCM_USE_AITER_CUSTOM_AR=1 pytest -v -s tests/compile/fusions_e2e/test_tp2_ar_rms.py -k 'test_tp2_ar_rms_fp8_fusions and qwen3 and ROCM_AITER_UNIFIED_ATTN and +quant_fp8 and (dynamo_partition or not +rms_norm)'
|
|
- pytest -v -s tests/compile/fusions_e2e/test_tp2_rocm_moe.py
|
|
|
|
- label: ":amd: (MI355) Fusion and Compile FP8"
|
|
timeout_in_minutes: 45
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx950nightly, amdmi355]
|
|
dind: false
|
|
agent_pool: mi355_2
|
|
num_gpus: 2
|
|
optional: true
|
|
working_dir: "/vllm-workspace/"
|
|
source_file_dependencies:
|
|
- vllm/compilation/
|
|
- vllm/model_executor/layers/
|
|
- vllm/v1/attention/
|
|
- vllm/distributed/
|
|
- vllm/_aiter_ops.py
|
|
- tests/compile/passes/
|
|
- tests/compile/correctness_e2e/test_attn_quant.py
|
|
- vllm/model_executor/kernels/linear/
|
|
- vllm/model_executor/models/llama.py
|
|
commands:
|
|
- amd-smi
|
|
- pytest -v -s tests/compile/passes/test_fusion_attn.py tests/compile/passes/test_silu_mul_quant_fusion.py tests/compile/passes/test_silu_mul_quant_manual_fusion.py
|
|
- pytest -v -s tests/compile/correctness_e2e/test_attn_quant.py
|
|
|
|
- label: ":amd: (MI355) Sequence Parallel Correctness"
|
|
timeout_in_minutes: 180
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx950nightly, amdmi355]
|
|
dind: false
|
|
agent_pool: mi355_2
|
|
num_gpus: 2
|
|
optional: true
|
|
working_dir: "/vllm-workspace/"
|
|
source_file_dependencies:
|
|
- .buildkite/test_areas/compile.yaml
|
|
- vllm/compilation/
|
|
- vllm/config/
|
|
- vllm/distributed/
|
|
- vllm/model_executor/kernels/linear/
|
|
- vllm/model_executor/layers/
|
|
- vllm/model_executor/models/llama.py
|
|
- vllm/v1/attention/
|
|
- vllm/v1/worker/
|
|
- vllm/v1/cudagraph_dispatcher.py
|
|
- vllm/platforms/rocm.py
|
|
- tests/compile/correctness_e2e/test_sequence_parallel.py
|
|
- tests/models/registry.py
|
|
- tests/utils.py
|
|
commands:
|
|
- export VLLM_TEST_CLEAN_GPU_MEMORY=1
|
|
- pytest -v -s tests/compile/correctness_e2e/test_sequence_parallel.py
|
|
|
|
- label: ":amd: (MI355) AsyncTP Correctness"
|
|
timeout_in_minutes: 60
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx950nightly, amdmi355]
|
|
dind: false
|
|
agent_pool: mi355_2
|
|
num_gpus: 2
|
|
optional: true
|
|
working_dir: "/vllm-workspace/"
|
|
source_file_dependencies:
|
|
- .buildkite/test_areas/compile.yaml
|
|
- vllm/compilation/
|
|
- vllm/config/
|
|
- vllm/distributed/
|
|
- vllm/model_executor/layers/
|
|
- vllm/model_executor/kernels/linear/
|
|
- vllm/model_executor/models/llama.py
|
|
- vllm/model_executor/models/qwen3.py
|
|
- vllm/v1/attention/
|
|
- vllm/_aiter_ops.py
|
|
- vllm/platforms/rocm.py
|
|
- vllm/transformers_utils/repo_utils.py
|
|
- tests/compile/correctness_e2e/test_async_tp.py
|
|
- tests/compile/fusions_e2e/common.py
|
|
- tests/models/registry.py
|
|
- tests/utils.py
|
|
commands:
|
|
- export VLLM_TEST_CLEAN_GPU_MEMORY=1
|
|
- pytest -v -s tests/compile/correctness_e2e/test_async_tp.py::test_rocm_async_tp_bf16_output_correctness
|
|
- pytest -v -s tests/compile/correctness_e2e/test_async_tp.py::test_async_tp_pass_correctness -k RedHatAI
|
|
- pytest -v -s tests/compile/correctness_e2e/test_async_tp.py::test_async_tp_pass_nvfp4_correctness
|
|
|
|
- label: ":amd: (MI355 DPX) Fusion E2E FP8 Config Sweep"
|
|
timeout_in_minutes: 45
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx950nightly, amdmi355]
|
|
dind: false
|
|
agent_pool: mi355_dpx
|
|
num_gpus: 1
|
|
optional: true
|
|
working_dir: "/vllm-workspace/"
|
|
source_file_dependencies:
|
|
- .buildkite/test_areas/compile.yaml
|
|
- csrc/quantization/
|
|
- vllm/compilation/
|
|
- vllm/model_executor/
|
|
- vllm/v1/attention/
|
|
- vllm/_aiter_ops.py
|
|
- vllm/platforms/rocm.py
|
|
- tests/compile/fusions_e2e/
|
|
commands:
|
|
- rocm-smi
|
|
- pytest -v -s tests/compile/fusions_e2e/test_tp1_quant.py -k "test_tp1_fp8_fusions and inductor_partition and not +rms_norm and not True and ((Llama-3 and TRITON_ATTN) or (Qwen3-30B and +quant_fp8 and TRITON_ATTN) or (DeepSeek-Coder and +quant_fp8 and TRITON_MLA))"
|
|
|
|
#-------------------------------------------------------- mi355 · distributed --------------------------------------------------------#
|
|
|
|
- label: ":amd: (MI355 DPX) EPLB Algorithm" # TBD
|
|
timeout_in_minutes: 30
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx950nightly, amdmi355]
|
|
dind: false
|
|
agent_pool: mi355_dpx
|
|
optional: true
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- vllm/distributed/eplb
|
|
- tests/distributed/test_eplb_algo.py
|
|
- tests/distributed/test_eplb_utils.py
|
|
- vllm/platforms/rocm.py
|
|
commands:
|
|
- pytest -v -s distributed/test_eplb_algo.py
|
|
- pytest -v -s distributed/test_eplb_utils.py
|
|
|
|
- label: ":amd: (MI355 DPX) Sharded RDT Weight Transfer"
|
|
timeout_in_minutes: 45
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx950nightly, amdmi355]
|
|
dind: false
|
|
agent_pool: mi355_dpx
|
|
num_gpus: 1
|
|
optional: true
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- vllm/distributed/weight_transfer/
|
|
- tests/distributed/test_sharded_rdt_plan.py
|
|
- tests/distributed/test_sharded_rdt_producer.py
|
|
- tests/distributed/test_sharded_rdt_trainer.py
|
|
- vllm/platforms/rocm.py
|
|
commands:
|
|
- export VLLM_WORKER_MULTIPROC_METHOD=spawn
|
|
- uv pip install --system --no-deps 'ray==2.56.1'
|
|
- pytest -v -s distributed/test_sharded_rdt_plan.py
|
|
- pytest -v -s distributed/test_sharded_rdt_producer.py
|
|
- pytest -v -s distributed/test_sharded_rdt_trainer.py
|
|
|
|
- label: ":amd: (MI355) GEMM-RS-AR"
|
|
timeout_in_minutes: 20
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx950nightly, amdmi355]
|
|
dind: false
|
|
agent_pool: mi355_4
|
|
num_gpus: 4
|
|
optional: true
|
|
working_dir: "/vllm-workspace/"
|
|
source_file_dependencies:
|
|
- .buildkite/test-amd.yaml
|
|
- tests/kernels/test_gemm_rs_ar.py
|
|
- tests/kernels/test_rocm_gemm_rs_ar.py
|
|
- tests/conftest.py
|
|
- vllm/models/common/ops/sequence_parallel.py
|
|
- vllm/model_executor/layers/linear.py
|
|
- vllm/model_executor/layers/utils.py
|
|
- vllm/model_executor/layers/quantization/online/
|
|
- vllm/model_executor/layers/quantization/utils/mxfp8_utils.py
|
|
- vllm/model_executor/kernels/linear/
|
|
- vllm/distributed/
|
|
- vllm/_aiter_ops.py
|
|
- vllm/_custom_ops.py
|
|
- csrc/rocm/skinny_gemms.cu
|
|
- vllm/platforms/rocm.py
|
|
commands:
|
|
- pytest -v -s tests/kernels/test_rocm_gemm_rs_ar.py
|
|
|
|
- label: ":amd: (MI355) Distributed DP Basic" # TBD
|
|
timeout_in_minutes: 45
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx950nightly, amdmi355]
|
|
dind: false
|
|
agent_pool: mi355_2
|
|
optional: true
|
|
num_gpus: 2
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- vllm/distributed/
|
|
- vllm/engine/
|
|
- vllm/executor/
|
|
- vllm/worker/worker_base.py
|
|
- vllm/v1/engine/
|
|
- vllm/v1/worker/
|
|
- tests/v1/distributed
|
|
- tests/entrypoints/launchers/api_server/test_multi_api_servers.py
|
|
- vllm/platforms/rocm.py
|
|
- tests/v1/e2e/general/test_sharded_sampling.py
|
|
- tests/v1/e2e/spec_decode/test_sharded_sampling.py
|
|
- tests/v1/e2e/spec_decode/utils.py
|
|
commands:
|
|
- export NCCL_CUMEM_HOST_ENABLE=0
|
|
- TP_SIZE=1 DP_SIZE=2 pytest -v -s v1/distributed/test_async_llm_dp.py
|
|
- TP_SIZE=1 DP_SIZE=2 pytest -v -s v1/distributed/test_eagle_dp.py
|
|
- TP_SIZE=1 DP_SIZE=2 pytest -v -s v1/distributed/test_external_lb_dp.py
|
|
- DP_SIZE=2 pytest -v -s entrypoints/launchers/api_server/test_multi_api_servers.py
|
|
- pytest -v -s v1/e2e/general/test_sharded_sampling.py
|
|
- pytest -v -s v1/e2e/spec_decode/test_sharded_sampling.py
|
|
|
|
- label: ":amd: (MI355) MoRI EP Healthy Baseline"
|
|
timeout_in_minutes: 10
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx950nightly, amdmi355]
|
|
dind: false
|
|
agent_pool: mi355_2
|
|
num_gpus: 2
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- csrc/moe/
|
|
- csrc/rocm/
|
|
- tests/kernels/moe/
|
|
- vllm/model_executor/layers/fused_moe/
|
|
- vllm/model_executor/layers/quantization/
|
|
- vllm/distributed/
|
|
- vllm/config/
|
|
- vllm/forward_context.py
|
|
- vllm/v1/worker/workspace.py
|
|
- vllm/utils/import_utils.py
|
|
- vllm/utils/math_utils.py
|
|
- vllm/utils/torch_utils.py
|
|
- vllm/platforms/
|
|
- vllm/_aiter_ops.py
|
|
optional: true
|
|
# TODO: Add fault injection, detection and recovery once MoRI supports them.
|
|
# This job currently checks healthy EP execution only.
|
|
commands:
|
|
- VLLM_WORKER_MULTIPROC_METHOD=spawn pytest -v -s -rs kernels/moe/test_moe_layer.py::test_moe_layer_mori_graph
|
|
- VLLM_TEST_ENABLE_MORI_MOE_LAYER=1 VLLM_WORKER_MULTIPROC_METHOD=spawn pytest -v -s -rs 'kernels/moe/test_moe_layer.py::test_moe_layer[False-mori_high_throughput-2-1-True]' --subtests='[32-512-512-8-2-bfloat16-None-False-False-False-False-mori_high_throughput-2-2-1]'
|
|
|
|
- label: ":amd: (MI355) NixlConnector PD + Spec Decode acceptance" # TBD
|
|
timeout_in_minutes: 45
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx950nightly, amdmi355]
|
|
dind: false
|
|
agent_pool: mi355_2
|
|
num_gpus: 2
|
|
optional: true
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- vllm/distributed/kv_transfer/kv_connector/v1/nixl/
|
|
- vllm/v1/worker/kv_connector_model_runner_mixin.py
|
|
- tests/v1/kv_connector/nixl_integration/
|
|
- vllm/platforms/rocm.py
|
|
commands:
|
|
- uv pip install --system -r /vllm-workspace/requirements/kv_connectors_rocm.txt
|
|
- KV_CACHE_MEMORY_BYTES=8G ATTENTION_BACKEND=TRITON_ATTN bash v1/kv_connector/nixl_integration/config_sweep_spec_decode_test.sh
|
|
|
|
- label: ":amd: (MI355) Distributed NixlConnector PD accuracy" # TBD
|
|
timeout_in_minutes: 60
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx950nightly, amdmi355]
|
|
dind: false
|
|
agent_pool: mi355_4
|
|
num_gpus: 4
|
|
optional: true
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- vllm/distributed/kv_transfer/kv_connector/v1/nixl/
|
|
- tests/v1/kv_connector/nixl_integration/
|
|
- vllm/platforms/rocm.py
|
|
commands:
|
|
- uv pip install --system -r /vllm-workspace/requirements/kv_connectors_rocm.txt
|
|
- ATTENTION_BACKEND=TRITON_ATTN bash v1/kv_connector/nixl_integration/config_sweep_accuracy_test.sh
|
|
|
|
- label: ":amd: (MI355) Distributed AITER NixlConnector PD accuracy"
|
|
timeout_in_minutes: 60
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx950nightly, amdmi355]
|
|
dind: false
|
|
agent_pool: mi355_4
|
|
num_gpus: 4
|
|
optional: true
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- vllm/distributed/kv_transfer/kv_connector/v1/nixl/
|
|
- tests/v1/kv_connector/nixl_integration/
|
|
- tests/v1/kv_connector/rocm_pd_accuracy_utils.py
|
|
- tests/utils.py
|
|
- vllm/v1/attention/backends/rocm_aiter_unified_attn.py
|
|
- vllm/platforms/rocm.py
|
|
- requirements/common.txt
|
|
- requirements/kv_connectors_rocm.txt
|
|
commands:
|
|
- uv pip install --system -r /vllm-workspace/requirements/kv_connectors_rocm.txt
|
|
- pytest -v -s v1/kv_connector/nixl_integration/test_accuracy_rocm.py
|
|
|
|
- label: ":amd: (MI355) Distributed MooncakeConnector HIP PD accuracy"
|
|
timeout_in_minutes: 30
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx950nightly, amdmi355]
|
|
dind: false
|
|
agent_pool: mi355_4
|
|
num_gpus: 4
|
|
optional: true
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- vllm/distributed/kv_transfer/kv_connector/v1/mooncake/
|
|
- tests/v1/kv_connector/mooncake_integration/
|
|
- tests/v1/kv_connector/rocm_pd_accuracy_utils.py
|
|
- tests/utils.py
|
|
- examples/disaggregated/mooncake_connector/
|
|
- vllm/v1/attention/backends/rocm_aiter_unified_attn.py
|
|
- vllm/platforms/rocm.py
|
|
- requirements/common.txt
|
|
- requirements/kv_connectors_rocm.txt
|
|
commands:
|
|
- uv pip install --system -r /vllm-workspace/requirements/kv_connectors_rocm.txt
|
|
- pytest -v -s v1/kv_connector/mooncake_integration/test_accuracy_rocm.py
|
|
|
|
- label: ":amd: (MI355) DP EP Distributed NixlConnector PD accuracy" # TBD
|
|
timeout_in_minutes: 40
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx950nightly, amdmi355]
|
|
dind: false
|
|
agent_pool: mi355_4
|
|
num_gpus: 4
|
|
optional: true
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- vllm/distributed/kv_transfer/kv_connector/v1/nixl/
|
|
- tests/v1/kv_connector/nixl_integration/
|
|
- vllm/platforms/rocm.py
|
|
commands:
|
|
- uv pip install --system -r /vllm-workspace/requirements/kv_connectors_rocm.txt
|
|
- DP_EP=1 ATTENTION_BACKEND=TRITON_ATTN bash v1/kv_connector/nixl_integration/config_sweep_accuracy_test.sh
|
|
|
|
- label: ":amd: (MI355) Kimi-Linear-48B-A3B Disaggregated DP EP"
|
|
timeout_in_minutes: 105
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx950nightly, amdmi355]
|
|
dind: false
|
|
agent_pool: mi355_4
|
|
num_gpus: 4
|
|
optional: true
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- vllm/platforms/rocm.py
|
|
- vllm/models/kimi_k3/
|
|
- vllm/distributed/kv_transfer/kv_connector/v1/nixl/
|
|
- tests/v1/kv_connector/nixl_integration/
|
|
- requirements/kv_connectors_rocm.txt
|
|
commands:
|
|
- export VLLM_WORKER_MULTIPROC_METHOD=spawn
|
|
- export ENABLE_HMA_FLAG='1'
|
|
- export DP_EP='1'
|
|
- export GPU_MEMORY_UTILIZATION='0.9'
|
|
- export VLLM_SSM_CONV_STATE_LAYOUT='DS'
|
|
- export PREFILLER_TP_SIZE='2'
|
|
- export DECODER_TP_SIZE='2'
|
|
- export MODEL_NAMES='moonshotai/Kimi-Linear-48B-A3B-Instruct'
|
|
- export PREFILL_BLOCK_SIZE='2048'
|
|
- export DECODE_BLOCK_SIZE='2048'
|
|
- export VLLM_SERVE_EXTRA_ARGS='--trust-remote-code,--mamba-cache-mode align,--max-model-len 32768'
|
|
- uv pip install --system -r /vllm-workspace/requirements/kv_connectors_rocm.txt
|
|
- bash v1/kv_connector/nixl_integration/run_accuracy_test.sh
|
|
|
|
- label: ":amd: (MI355) Distributed Features" # TBD
|
|
timeout_in_minutes: 90
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx950nightly, amdmi355]
|
|
dind: false
|
|
agent_pool: mi355_2
|
|
num_gpus: 2
|
|
optional: true
|
|
working_dir: "/vllm-workspace/"
|
|
source_file_dependencies:
|
|
- csrc/custom_quickreduce.cu
|
|
- csrc/ops.h
|
|
- csrc/torch_bindings.cpp
|
|
- vllm/distributed/
|
|
- vllm/model_executor/layers/
|
|
- vllm/entrypoints/llm.py
|
|
- vllm/config/parallel.py
|
|
- vllm/model_executor/layers/fused_moe/
|
|
- vllm/v1/engine/
|
|
- vllm/v1/executor/
|
|
- vllm/v1/worker/
|
|
- vllm/v1/distributed/
|
|
- vllm/v1/attention/backends/
|
|
- vllm/v1/attention/selector.py
|
|
- vllm/_aiter_ops.py
|
|
- vllm/_custom_ops.py
|
|
- vllm/platforms/rocm.py
|
|
- vllm/envs.py
|
|
- examples/offline_inference/data_parallel.py
|
|
- tests/distributed/test_context_parallel.py
|
|
- tests/distributed/test_rocm_aiter_custom_ar.py
|
|
- tests/distributed/test_rocm_quick_reduce.py
|
|
- tests/distributed/test_quick_all_reduce.py
|
|
- tests/v1/e2e/general/test_rocm_aiter_custom_ar.py
|
|
- tests/v1/distributed/test_dbo.py
|
|
- tests/utils.py
|
|
- examples/features/data_parallel/data_parallel_offline.py
|
|
- examples/rl/rlhf_async_new_apis.py
|
|
- tests/distributed/test_weight_transfer.py
|
|
- tests/distributed/test_packed_tensor.py
|
|
commands:
|
|
- pytest -v -s tests/distributed/test_context_parallel.py
|
|
- VLLM_ALLOW_INSECURE_SERIALIZATION=1 python3 examples/rl/rlhf_async_new_apis.py
|
|
- VLLM_LOGGING_LEVEL=DEBUG python3 examples/features/data_parallel/data_parallel_offline.py --model=Qwen/Qwen1.5-MoE-A2.7B -tp=1 -dp=2 --max-model-len=2048 --all2all-backend=deepep_high_throughput
|
|
- VLLM_LOGGING_LEVEL=DEBUG python3 examples/features/data_parallel/data_parallel_offline.py --model=Qwen/Qwen1.5-MoE-A2.7B -tp=1 -dp=2 --max-model-len=2048 --all2all-backend=allgather_reducescatter --disable-nccl-for-dp-synchronization
|
|
- pytest -v -s tests/v1/distributed/test_dbo.py
|
|
- VLLM_ALLOW_INSECURE_SERIALIZATION=1 pytest -v -s tests/distributed/test_weight_transfer.py
|
|
- pytest -v -s tests/distributed/test_packed_tensor.py
|
|
- pytest -v -s tests/distributed/test_rocm_aiter_custom_ar.py
|
|
- pytest -v -s tests/v1/e2e/general/test_rocm_aiter_custom_ar.py
|
|
- pytest -v -s tests/distributed/test_rocm_quick_reduce.py
|
|
- pytest -v -s tests/distributed/test_quick_all_reduce.py
|
|
|
|
- label: ":amd: (MI355) Distributed AgRs All2All"
|
|
timeout_in_minutes: 30
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx950nightly, amdmi355]
|
|
dind: false
|
|
agent_pool: mi355_2
|
|
num_gpus: 2
|
|
optional: true
|
|
working_dir: "/vllm-workspace/"
|
|
source_file_dependencies:
|
|
- .buildkite/test_areas/distributed.yaml
|
|
- vllm/distributed/
|
|
- vllm/config/parallel.py
|
|
- vllm/config/vllm.py
|
|
- vllm/forward_context.py
|
|
- vllm/platforms/rocm.py
|
|
- tests/distributed/test_mnnvl_alltoall.py
|
|
- tests/utils.py
|
|
commands:
|
|
- pytest -v -s tests/distributed/test_mnnvl_alltoall.py::test_args_dispatch_combine
|
|
|
|
- label: ":amd: (MI355) Fusion E2E TP2 Quick" # TBD
|
|
timeout_in_minutes: 60
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx950nightly, amdmi355]
|
|
dind: false
|
|
agent_pool: mi355_2
|
|
num_gpus: 2
|
|
optional: true
|
|
working_dir: "/vllm-workspace/"
|
|
source_file_dependencies:
|
|
- csrc/quantization/
|
|
- vllm/compilation/
|
|
- vllm/distributed/
|
|
- vllm/model_executor/
|
|
- vllm/v1/attention/
|
|
- tests/compile/fusions_e2e/
|
|
- vllm/_aiter_ops.py
|
|
- vllm/platforms/rocm.py
|
|
commands:
|
|
- export VLLM_TEST_CLEAN_GPU_MEMORY=1
|
|
- >-
|
|
VLLM_ROCM_USE_AITER=1
|
|
VLLM_ROCM_USE_AITER_CUSTOM_AR=1
|
|
pytest -v -s tests/compile/fusions_e2e/test_tp2_ar_rms.py
|
|
-k 'test_tp2_ar_rms_fp8_fusions and inductor_partition and
|
|
not +quant_fp8 and not +rms_norm and Llama-3 and
|
|
ROCM_AITER_UNIFIED_ATTN'
|
|
- >-
|
|
VLLM_ROCM_USE_AITER=1
|
|
VLLM_ROCM_USE_AITER_CUSTOM_AR=1
|
|
pytest -v -s tests/compile/fusions_e2e/test_tp2_ar_rms.py
|
|
-k 'test_tp2_ar_rms_fp8_fusions and inductor_partition and
|
|
+quant_fp8 and +rms_norm and qwen3 and
|
|
ROCM_AITER_UNIFIED_ATTN'
|
|
- >-
|
|
VLLM_ROCM_USE_AITER=1
|
|
VLLM_ROCM_USE_AITER_CUSTOM_AR=1
|
|
pytest -v -s tests/compile/fusions_e2e/test_tp2_ar_rms.py
|
|
-k 'test_tp2_ar_rms_fp8_fusions and inductor_partition and
|
|
not +quant_fp8 and not +rms_norm and DeepSeek-Coder and
|
|
ROCM_AITER_MLA'
|
|
|
|
- label: ":amd: (MI355) DSv4-Flash Disaggregated DP EP" # TBD
|
|
timeout_in_minutes: 85
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx950nightly, amdmi355]
|
|
dind: false
|
|
agent_pool: mi355_8
|
|
num_gpus: 8
|
|
optional: true
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- requirements/kv_connectors_rocm.txt
|
|
- vllm/distributed/kv_transfer/kv_connector/v1/nixl/
|
|
- vllm/model_executor/layers/fused_moe/
|
|
- vllm/model_executor/layers/quantization/
|
|
- vllm/models/deepseek_v4/
|
|
- vllm/v1/attention/
|
|
- tests/v1/kv_connector/nixl_integration/
|
|
- vllm/_aiter_ops.py
|
|
- vllm/platforms/rocm.py
|
|
commands:
|
|
- uv pip install --system -r /vllm-workspace/requirements/kv_connectors_rocm.txt
|
|
# A mirror does not inherit the parent env, so it is repeated here in
|
|
# the parent's order. The layouts must stay empty: unset, the script
|
|
# defaults to HND, which the V4 indexer rejects.
|
|
- >-
|
|
VLLM_ROCM_USE_AITER=1
|
|
ENABLE_HMA_FLAG=1
|
|
DP_EP=1
|
|
GPU_MEMORY_UTILIZATION=0.85
|
|
PREFILLER_TP_SIZE=4
|
|
DECODER_TP_SIZE=4
|
|
PREFILL_BLOCK_SIZE=256
|
|
DECODE_BLOCK_SIZE=256
|
|
MODEL_NAMES=deepseek-ai/DeepSeek-V4-Flash
|
|
VLLM_ENGINE_READY_TIMEOUT_S=1800
|
|
VLLM_SERVE_EXTRA_ARGS=--trust-remote-code,--kv-cache-dtype,fp8
|
|
PREFILLER_KV_LAYOUT=
|
|
DECODER_KV_LAYOUT=
|
|
bash v1/kv_connector/nixl_integration/run_accuracy_test.sh
|
|
|
|
#---------------------------------------------------------- mi355 · engine -----------------------------------------------------------#
|
|
|
|
- label: ":amd: (MI355 DPX) Engine Core" # TBD
|
|
timeout_in_minutes: 40
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx950nightly, amdmi355]
|
|
dind: false
|
|
agent_pool: mi355_dpx
|
|
optional: true
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- vllm/
|
|
- tests/engine
|
|
- tests/test_sequence
|
|
- tests/test_config
|
|
- tests/test_logger
|
|
- tests/test_vllm_port
|
|
- tests/jit_monitor/test_hooks.py
|
|
- tests/jit_monitor/test_hooks_gpu.py
|
|
- vllm/compilation/
|
|
- vllm/config/
|
|
- vllm/engine/
|
|
- vllm/entrypoints/logger.py
|
|
- vllm/envs.py
|
|
- vllm/logger.py
|
|
- vllm/logging_utils/
|
|
- vllm/platforms/
|
|
- vllm/sequence.py
|
|
- vllm/triton_utils/
|
|
- vllm/utils/
|
|
commands:
|
|
- pytest -v -s engine test_sequence.py test_config.py test_logger.py test_vllm_port.py jit_monitor/test_hooks.py jit_monitor/test_hooks_gpu.py
|
|
|
|
#-------------------------------------------------------- mi355 · entrypoints --------------------------------------------------------#
|
|
|
|
- label: ":amd: (MI355 DPX) Entrypoints Integration (LLM)" # TBD
|
|
timeout_in_minutes: 90
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx950nightly, amdmi355]
|
|
dind: false
|
|
agent_pool: mi355_dpx
|
|
optional: true
|
|
fast_check: true
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- vllm/
|
|
- tests/entrypoints/llm
|
|
commands:
|
|
- export VLLM_WORKER_MULTIPROC_METHOD=spawn
|
|
- pytest -v -s entrypoints/llm --ignore=entrypoints/llm/test_generate.py --ignore=entrypoints/llm/test_collective_rpc.py --ignore=entrypoints/llm/offline_mode
|
|
- pytest -v -s entrypoints/llm/test_generate.py # it needs a clean process
|
|
- pytest -v -s entrypoints/llm/offline_mode # Needs to avoid interference with other tests
|
|
|
|
- label: ":amd: (MI355 DPX) Entrypoints Integration (API Server) %N"
|
|
timeout_in_minutes: 65
|
|
parallelism: 4
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx950nightly, amdmi355]
|
|
dind: false
|
|
agent_pool: mi355_dpx
|
|
optional: true
|
|
fast_check: true
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- vllm/
|
|
- tests/entrypoints/serve
|
|
- tests/entrypoints/scale_out
|
|
commands:
|
|
- export VLLM_WORKER_MULTIPROC_METHOD=spawn
|
|
- pytest -v -s entrypoints/serve --ignore=entrypoints/serve/dev/rpc --num-shards=$$BUILDKITE_PARALLEL_JOB_COUNT --shard-id=$$BUILDKITE_PARALLEL_JOB
|
|
- if [ "$$BUILDKITE_PARALLEL_JOB" = "1" ]; then PYTHONPATH=/vllm-workspace pytest -v -s entrypoints/serve/dev/rpc; fi
|
|
- pytest -v -s entrypoints/scale_out --num-shards=$$BUILDKITE_PARALLEL_JOB_COUNT --shard-id=$$BUILDKITE_PARALLEL_JOB
|
|
|
|
- label: ":amd: (MI355 DPX) Entrypoints Integration (OpenAI API completion)" # TBD
|
|
timeout_in_minutes: 65
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx950nightly, amdmi355]
|
|
dind: false
|
|
agent_pool: mi355_dpx
|
|
fast_check: true
|
|
optional: true
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- vllm/
|
|
- tests/entrypoints/openai
|
|
commands:
|
|
- export VLLM_WORKER_MULTIPROC_METHOD=spawn
|
|
- pytest -v -s entrypoints/openai/completion --ignore=entrypoints/openai/completion/test_tensorizer_entrypoint.py
|
|
- pytest -v -s entrypoints/openai --ignore=entrypoints/openai/completion --ignore=entrypoints/openai/chat_completion --ignore=entrypoints/openai/responses --ignore=entrypoints/openai/correctness
|
|
|
|
- label: ":amd: (MI355 DPX) Entrypoints Integration (OpenAI API chat_completion)" # TBD
|
|
timeout_in_minutes: 70
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx950nightly, amdmi355]
|
|
dind: false
|
|
agent_pool: mi355_dpx
|
|
fast_check: true
|
|
optional: true
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- vllm/
|
|
- tests/entrypoints/openai
|
|
commands:
|
|
- export VLLM_WORKER_MULTIPROC_METHOD=spawn
|
|
- pytest -v -s entrypoints/openai/chat_completion
|
|
|
|
- label: ":amd: (MI355 DPX) Entrypoints Integration (API Server Generate)" # TBD
|
|
timeout_in_minutes: 65
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx950nightly, amdmi355]
|
|
dind: false
|
|
agent_pool: mi355_dpx
|
|
optional: true
|
|
fast_check: true
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- vllm/
|
|
- tests/tool_use
|
|
- tests/entrypoints/tool_parsers
|
|
- tests/entrypoints/anthropic
|
|
- tests/entrypoints/cohere
|
|
- tests/entrypoints/generate
|
|
commands:
|
|
- pytest -v -s tool_use
|
|
- pytest -v -s entrypoints/tool_parsers
|
|
- pytest -v -s entrypoints/generate
|
|
- pytest -v -s entrypoints/anthropic
|
|
- pytest -v -s entrypoints/cohere
|
|
|
|
- label: ":amd: (MI355 DPX) OpenAI API Correctness" # TBD
|
|
timeout_in_minutes: 30
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx950nightly, amdmi355]
|
|
dind: false
|
|
agent_pool: mi355_dpx
|
|
optional: true
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- csrc/
|
|
- vllm/entrypoints/openai/
|
|
- vllm/model_executor/layers/
|
|
- vllm/v1/attention/backends/
|
|
- vllm/v1/attention/selector.py
|
|
- vllm/_aiter_ops.py
|
|
- vllm/platforms/rocm.py
|
|
- vllm/model_executor/model_loader/
|
|
commands:
|
|
- bash ../tools/install_torchcodec_rocm.sh || exit 1
|
|
- pytest -s entrypoints/openai/correctness/
|
|
|
|
- label: ":amd: (MI355 DPX) Entrypoints Integration (Speech to Text)" # TBD
|
|
timeout_in_minutes: 60
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx950nightly, amdmi355]
|
|
dind: false
|
|
agent_pool: mi355_dpx
|
|
fast_check: true
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- vllm/
|
|
- tests/entrypoints/speech_to_text
|
|
optional: true
|
|
commands:
|
|
- export VLLM_WORKER_MULTIPROC_METHOD=spawn
|
|
- pytest -v -s entrypoints/speech_to_text
|
|
|
|
- label: ":amd: (MI355 DPX) Entrypoints Integration (Multimodal)"
|
|
timeout_in_minutes: 55
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx950nightly, amdmi355]
|
|
dind: false
|
|
agent_pool: mi355_dpx
|
|
fast_check: true
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- vllm/
|
|
- tests/entrypoints/multimodal
|
|
optional: true
|
|
commands:
|
|
- export VLLM_WORKER_MULTIPROC_METHOD=spawn
|
|
- pytest -v -s entrypoints/multimodal
|
|
|
|
- label: ":amd: (MI355 DPX) Entrypoints Integration (Pooling) Shard %N" # TBD
|
|
timeout_in_minutes: 65
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx950nightly, amdmi355]
|
|
dind: false
|
|
agent_pool: mi355_dpx
|
|
fast_check: true
|
|
parallelism: 4
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- vllm/
|
|
- tests/entrypoints/pooling
|
|
optional: true
|
|
commands:
|
|
- export VLLM_WORKER_MULTIPROC_METHOD=spawn
|
|
- pytest -v -s entrypoints/pooling --num-shards=$$BUILDKITE_PARALLEL_JOB_COUNT --shard-id=$$BUILDKITE_PARALLEL_JOB
|
|
|
|
#----------------------------------------------------------- mi355 · evals -----------------------------------------------------------#
|
|
|
|
- label: ":amd: (MI355 DPX) Multimodal Accuracy Eval (Small Models)" # TBD
|
|
timeout_in_minutes: 35
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx950nightly, amdmi355]
|
|
dind: false
|
|
agent_pool: mi355_dpx
|
|
optional: true
|
|
working_dir: "/vllm-workspace/.buildkite/lm-eval-harness"
|
|
source_file_dependencies:
|
|
- vllm/multimodal/
|
|
- vllm/inputs/
|
|
- vllm/v1/core/
|
|
- vllm/platforms/rocm.py
|
|
- vllm/model_executor/model_loader/
|
|
commands:
|
|
- pytest -s -v test_lm_eval_correctness.py --config-list-file=configs/models-mm-small.txt --tp-size=1
|
|
|
|
- label: ":amd: (MI355) GPQA Eval (GPT-OSS)" # TBD
|
|
timeout_in_minutes: 40
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx950nightly, amdmi355]
|
|
dind: false
|
|
agent_pool: mi355_2
|
|
num_gpus: 2
|
|
optional: true
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- csrc/
|
|
- vllm/model_executor/layers/quantization
|
|
- vllm/model_executor/models/
|
|
- vllm/model_executor/model_loader/
|
|
- vllm/v1/attention/backends/
|
|
- vllm/v1/attention/selector.py
|
|
- vllm/model_executor/layers/fused_moe/
|
|
- tests/evals/gpt_oss/
|
|
- vllm/_aiter_ops.py
|
|
- vllm/platforms/rocm.py
|
|
commands:
|
|
- pytest -s -v evals/gpt_oss/test_gpqa_correctness.py --config-list-file=configs/models-gfx950.txt
|
|
|
|
- label: ":amd: (MI355 DPX) LM Eval Small Models"
|
|
timeout_in_minutes: 45
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx950nightly, amdmi355]
|
|
dind: false
|
|
agent_pool: mi355_dpx
|
|
num_gpus: 1
|
|
optional: true
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- .buildkite/test_areas/lm_eval.yaml
|
|
- csrc/
|
|
- tests/evals/gsm8k/
|
|
- vllm/model_executor/layers/quantization
|
|
- vllm/model_executor/kernels/linear/
|
|
- vllm/model_executor/layers/fused_moe/
|
|
- vllm/model_executor/models/
|
|
- vllm/model_executor/model_loader/
|
|
- vllm/v1/attention/backends/
|
|
- vllm/v1/attention/selector.py
|
|
- vllm/_aiter_ops.py
|
|
- vllm/platforms/rocm.py
|
|
commands:
|
|
- pytest -s -v evals/gsm8k/test_gsm8k_correctness.py --config-list-file=configs/models-gfx950-small.txt
|
|
|
|
- label: ":amd: (MI355 DPX) LM Eval Watermarking"
|
|
timeout_in_minutes: 45
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx950nightly, amdmi355]
|
|
dind: false
|
|
agent_pool: mi355_dpx
|
|
num_gpus: 1
|
|
optional: true
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- .buildkite/test_areas/lm_eval.yaml
|
|
- tests/evals/gsm8k/
|
|
- vllm/config/watermarking.py
|
|
- vllm/sampling_params.py
|
|
- vllm/v1/watermarking/
|
|
- vllm/v1/worker/gpu/
|
|
- vllm/platforms/rocm.py
|
|
commands:
|
|
- export VLLM_WORKER_MULTIPROC_METHOD=spawn
|
|
- pytest -s -v evals/gsm8k/test_gsm8k_correctness.py --config-list-file=configs/models-watermark.txt
|
|
|
|
- label: ":amd: (MI355) LM Eval Watermark Feature Combination"
|
|
timeout_in_minutes: 60
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx950nightly, amdmi355]
|
|
dind: false
|
|
agent_pool: mi355_2
|
|
num_gpus: 2
|
|
optional: true
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- .buildkite/test_areas/lm_eval.yaml
|
|
- tests/evals/gsm8k/
|
|
- vllm/config/speculative.py
|
|
- vllm/config/vllm.py
|
|
- vllm/config/watermarking.py
|
|
- vllm/sampling_params.py
|
|
- vllm/model_executor/models/qwen3_dspark.py
|
|
- vllm/model_executor/models/qwen3_dflash.py
|
|
- vllm/model_executor/warmup/
|
|
- vllm/v1/watermarking/
|
|
- vllm/v1/worker/gpu/
|
|
- vllm/platforms/rocm.py
|
|
commands:
|
|
- export VLLM_WORKER_MULTIPROC_METHOD=spawn
|
|
- pytest -s -v evals/gsm8k/test_gsm8k_correctness.py --config-list-file=configs/models-watermark-dspark-tp2.txt
|
|
|
|
- label: ":amd: (MI355) LM Eval Qwen3-5 Models" # TBD
|
|
timeout_in_minutes: 75
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx950nightly, amdmi355]
|
|
dind: false
|
|
agent_pool: mi355_2
|
|
num_gpus: 2
|
|
optional: true
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- vllm/model_executor/models/qwen3_5.py
|
|
- vllm/model_executor/models/qwen3_5_mtp.py
|
|
- vllm/transformers_utils/configs/qwen3_5.py
|
|
- vllm/transformers_utils/configs/qwen3_5_moe.py
|
|
- vllm/model_executor/models/qwen2.py
|
|
- vllm/model_executor/models/qwen3.py
|
|
- vllm/model_executor/models/qwen3_next.py
|
|
- vllm/model_executor/models/qwen3_next_mtp.py
|
|
- vllm/distributed/
|
|
- vllm/model_executor/layers/fused_moe/
|
|
- vllm/third_party/flash_linear_attention/ops/
|
|
- vllm/model_executor/layers/quantization/modelopt.py
|
|
- vllm/model_executor/kernels/linear/nvfp4/
|
|
- vllm/v1/attention/
|
|
- vllm/v1/worker/
|
|
- tests/evals/gsm8k/
|
|
- tests/utils.py
|
|
- vllm/_aiter_ops.py
|
|
- vllm/platforms/rocm.py
|
|
commands:
|
|
- pytest -s -v evals/gsm8k/test_gsm8k_correctness.py --config-list-file=configs/models-qwen35-mi355.txt
|
|
|
|
- label: ":amd: (MI355) LM Eval Large Models EP"
|
|
timeout_in_minutes: 120
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx950nightly, amdmi355]
|
|
dind: false
|
|
agent_pool: mi355_2
|
|
num_gpus: 2
|
|
optional: true
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- .buildkite/test_areas/lm_eval.yaml
|
|
- tests/evals/gsm8k/
|
|
- tests/utils.py
|
|
- tests/conftest.py
|
|
- csrc/
|
|
- vllm/config/
|
|
- vllm/distributed/
|
|
- vllm/model_executor/models/config.py
|
|
- vllm/model_executor/models/registry.py
|
|
- vllm/model_executor/models/utils.py
|
|
- vllm/model_executor/models/qwen3_next.py
|
|
- vllm/model_executor/models/qwen3_next_mtp.py
|
|
- vllm/model_executor/models/nemotron_h.py
|
|
- vllm/model_executor/models/nemotron_h_mtp.py
|
|
- vllm/transformers_utils/configs/qwen3_next.py
|
|
- vllm/transformers_utils/configs/nemotron_h.py
|
|
- vllm/model_executor/model_loader/
|
|
- vllm/model_executor/layers/quantization/
|
|
- vllm/model_executor/layers/fused_moe/
|
|
- vllm/model_executor/layers/mamba/
|
|
- vllm/model_executor/layers/layernorm.py
|
|
- vllm/model_executor/layers/linear.py
|
|
- vllm/model_executor/kernels/linear/
|
|
- vllm/third_party/flash_linear_attention/ops/
|
|
- vllm/v1/attention/
|
|
- vllm/v1/spec_decode/
|
|
- vllm/v1/worker/
|
|
- vllm/_aiter_ops.py
|
|
- vllm/platforms/rocm.py
|
|
commands:
|
|
- export VLLM_WORKER_MULTIPROC_METHOD=spawn
|
|
- pytest -s -v evals/gsm8k/test_gsm8k_correctness.py --config-list-file=configs/models-blackwell-ep-mi355.txt
|
|
|
|
- label: ":amd: (MI355) Qwen3-8-Flash-Next-FP8 Accuracy Eval"
|
|
timeout_in_minutes: 105
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx950nightly, amdmi355]
|
|
dind: false
|
|
agent_pool: mi355_4
|
|
num_gpus: 4
|
|
optional: true
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- csrc/libtorch_stable/gdn/
|
|
- tests/evals/qwen4_exp/
|
|
- vllm/model_executor/layers/mamba/
|
|
- vllm/models/qwen4_exp/
|
|
- vllm/transformers_utils/configs/qwen4_exp.py
|
|
- vllm/v1/attention/backends/short_conv_attn.py
|
|
- vllm/platforms/rocm.py
|
|
commands:
|
|
- export VLLM_WORKER_MULTIPROC_METHOD=spawn
|
|
- uv pip install --system 'evalscope==1.10.0'
|
|
- pytest -s -v evals/qwen4_exp/test_accuracy.py --config-list-file=configs/models-rocm.txt
|
|
|
|
- label: ":amd: (MI355 DPX) LM Eval TurboQuant k8v4"
|
|
timeout_in_minutes: 30
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx950nightly, amdmi355]
|
|
dind: false
|
|
agent_pool: mi355_dpx
|
|
num_gpus: 1
|
|
optional: true
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- tests/evals/gsm8k/
|
|
- tests/utils.py
|
|
- tests/conftest.py
|
|
- vllm/model_executor/layers/quantization/turboquant/
|
|
- vllm/v1/attention/backends/turboquant_attn.py
|
|
- vllm/v1/attention/backends/triton_attn.py
|
|
- vllm/v1/attention/ops/triton_unified_attention.py
|
|
- vllm/v1/attention/selector.py
|
|
- vllm/v1/attention/ops/triton_turboquant_decode.py
|
|
- vllm/v1/attention/ops/triton_turboquant_store.py
|
|
- vllm/v1/attention/ops/flydsl_turboquant_decode.py
|
|
- vllm/v1/attention/ops/flydsl_kernels/
|
|
- vllm/v1/attention/ops/turboquant_soa/
|
|
- vllm/platforms/rocm.py
|
|
commands:
|
|
- pytest -s -v evals/gsm8k/test_gsm8k_correctness.py --config-list-file=evals/gsm8k/configs/models-turboquant-k8v4-rocm.txt
|
|
|
|
- label: ":amd: (MI355 DPX) LM Eval TurboQuant t4nc"
|
|
timeout_in_minutes: 30
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx950nightly, amdmi355]
|
|
dind: false
|
|
agent_pool: mi355_dpx
|
|
num_gpus: 1
|
|
optional: true
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- tests/evals/gsm8k/
|
|
- tests/utils.py
|
|
- tests/conftest.py
|
|
- vllm/model_executor/layers/quantization/turboquant/
|
|
- vllm/v1/attention/backends/turboquant_attn.py
|
|
- vllm/v1/attention/backends/triton_attn.py
|
|
- vllm/v1/attention/ops/triton_unified_attention.py
|
|
- vllm/v1/attention/selector.py
|
|
- vllm/v1/attention/ops/triton_turboquant_decode.py
|
|
- vllm/v1/attention/ops/triton_turboquant_store.py
|
|
- vllm/v1/attention/ops/flydsl_turboquant_decode.py
|
|
- vllm/v1/attention/ops/flydsl_kernels/
|
|
- vllm/v1/attention/ops/turboquant_soa/
|
|
- vllm/platforms/rocm.py
|
|
commands:
|
|
- pytest -s -v evals/gsm8k/test_gsm8k_correctness.py --config-list-file=evals/gsm8k/configs/models-turboquant-t4nc-rocm.txt
|
|
|
|
- label: ":amd: (MI355 DPX) LM Eval TurboQuant k3v4nc"
|
|
timeout_in_minutes: 30
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx950nightly, amdmi355]
|
|
dind: false
|
|
agent_pool: mi355_dpx
|
|
num_gpus: 1
|
|
optional: true
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- tests/evals/gsm8k/
|
|
- tests/utils.py
|
|
- tests/conftest.py
|
|
- vllm/model_executor/layers/quantization/turboquant/
|
|
- vllm/v1/attention/backends/turboquant_attn.py
|
|
- vllm/v1/attention/backends/triton_attn.py
|
|
- vllm/v1/attention/ops/triton_unified_attention.py
|
|
- vllm/v1/attention/selector.py
|
|
- vllm/v1/attention/ops/triton_turboquant_decode.py
|
|
- vllm/v1/attention/ops/triton_turboquant_store.py
|
|
- vllm/v1/attention/ops/flydsl_turboquant_decode.py
|
|
- vllm/v1/attention/ops/flydsl_kernels/
|
|
- vllm/v1/attention/ops/turboquant_soa/
|
|
- vllm/platforms/rocm.py
|
|
commands:
|
|
- pytest -s -v evals/gsm8k/test_gsm8k_correctness.py --config-list-file=evals/gsm8k/configs/models-turboquant-k3v4nc-rocm.txt
|
|
|
|
- label: ":amd: (MI355 DPX) LM Eval TurboQuant t3nc"
|
|
timeout_in_minutes: 30
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx950nightly, amdmi355]
|
|
dind: false
|
|
agent_pool: mi355_dpx
|
|
num_gpus: 1
|
|
optional: true
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- .buildkite/test-amd.yaml
|
|
- tests/evals/gsm8k/
|
|
- tests/utils.py
|
|
- tests/conftest.py
|
|
- vllm/model_executor/layers/quantization/turboquant/
|
|
- vllm/v1/attention/backends/turboquant_attn.py
|
|
- vllm/v1/attention/backends/triton_attn.py
|
|
- vllm/v1/attention/ops/triton_unified_attention.py
|
|
- vllm/v1/attention/selector.py
|
|
- vllm/v1/attention/ops/triton_turboquant_decode.py
|
|
- vllm/v1/attention/ops/triton_turboquant_store.py
|
|
- vllm/v1/attention/ops/flydsl_turboquant_decode.py
|
|
- vllm/v1/attention/ops/flydsl_kernels/
|
|
- vllm/v1/attention/ops/turboquant_soa/
|
|
- vllm/platforms/rocm.py
|
|
commands:
|
|
- pytest -s -v evals/gsm8k/test_gsm8k_correctness.py --config-list-file=evals/gsm8k/configs/models-turboquant-t3nc-rocm.txt
|
|
|
|
- label: ":amd: (MI355) LM Eval Spec Decode (Fixed-length FP8)" # TBD
|
|
timeout_in_minutes: 145
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx950nightly, amdmi355]
|
|
dind: false
|
|
agent_pool: mi355_4
|
|
num_gpus: 4
|
|
optional: true
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- tests/evals/gsm8k/
|
|
- tests/utils.py
|
|
- vllm/distributed/
|
|
- vllm/models/deepseek_v4/
|
|
- vllm/model_executor/models/qwen3_dspark.py
|
|
- vllm/model_executor/layers/fused_moe/
|
|
- vllm/v1/spec_decode/metrics.py
|
|
- vllm/v1/worker/gpu/spec_decode/dspark/
|
|
- vllm/v1/worker/gpu/spec_decode/rejection_sampler.py
|
|
- vllm/v1/worker/gpu/spec_decode/speculator.py
|
|
- vllm/model_executor/layers/quantization/
|
|
- vllm/v1/attention/
|
|
- vllm/_aiter_ops.py
|
|
- vllm/platforms/rocm.py
|
|
commands:
|
|
- pytest -s -v evals/gsm8k/test_gsm8k_correctness.py --config-list-file=configs/models-spec-decode-rocm.txt
|
|
|
|
- label: ':amd: (MI355) LM Eval Small Models Distributed'
|
|
timeout_in_minutes: 145
|
|
mirror_hardwares:
|
|
- amdexperimental
|
|
- amdproduction
|
|
- amdgfx950nightly
|
|
- amdmi355
|
|
dind: false
|
|
agent_pool: mi355_2
|
|
num_gpus: 2
|
|
optional: true
|
|
working_dir: /vllm-workspace/tests
|
|
source_file_dependencies:
|
|
- csrc/
|
|
- vllm/model_executor/layers/quantization
|
|
- vllm/model_executor/models/
|
|
- vllm/model_executor/model_loader/
|
|
- vllm/v1/attention/backends/
|
|
- vllm/v1/attention/selector.py
|
|
- vllm/v1/worker/
|
|
- vllm/v1/core/
|
|
- vllm/config/
|
|
- tests/evals/gsm8k/configs/models-mi3xx-fp8-and-mixed.txt
|
|
- vllm/_aiter_ops.py
|
|
- vllm/platforms/rocm.py
|
|
- tests/evals/gsm8k/
|
|
- tests/utils.py
|
|
- vllm/distributed/
|
|
- vllm/transformers_utils/configs/diffusion_gemma.py
|
|
- vllm/v1/attention/ops/
|
|
- tests/evals/gsm8k/configs/models-small-tp-mi355.txt
|
|
commands:
|
|
- pytest -s -v evals/gsm8k/test_gsm8k_correctness.py --config-list-file=configs/models-mi3xx-fp8-and-mixed.txt
|
|
- pytest -s -v evals/gsm8k/test_gsm8k_correctness.py --config-list-file=configs/models-small-tp.txt
|
|
- pytest -s -v evals/gsm8k/test_gsm8k_correctness.py --config-list-file=configs/models-small-tp-mi355.txt
|
|
|
|
- label: ":amd: (MI355) MoE Refactor Integration Shard %N"
|
|
timeout_in_minutes: 105
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx950nightly, amdmi355]
|
|
dind: false
|
|
agent_pool: mi355_2
|
|
num_gpus: 2
|
|
parallelism: 4
|
|
optional: true
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- vllm/platforms/rocm.py
|
|
- tests/evals/gsm8k/
|
|
- vllm/model_executor/layers/fused_moe/
|
|
- vllm/model_executor/layers/quantization/
|
|
- vllm/_aiter_ops.py
|
|
commands:
|
|
- export VLLM_WORKER_MULTIPROC_METHOD=spawn
|
|
- pytest -s -v evals/gsm8k/test_gsm8k_correctness.py --config-list-file=evals/gsm8k/configs/moe-refactor/config-rocm-shard-$$BUILDKITE_PARALLEL_JOB.txt
|
|
|
|
|
|
- label: ":amd: (MI355) Qwen3-30B-A3B-FP8-block Sync EPLB Accuracy" # TBD
|
|
timeout_in_minutes: 25
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx950nightly, amdmi355]
|
|
dind: false
|
|
agent_pool: mi355_2
|
|
num_gpus: 2
|
|
optional: true
|
|
working_dir: "/vllm-workspace"
|
|
source_file_dependencies:
|
|
- vllm/model_executor/models/
|
|
- vllm/model_executor/model_loader/
|
|
- vllm/model_executor/layers/quantization/
|
|
- vllm/model_executor/layers/fused_moe/
|
|
- vllm/distributed/eplb
|
|
- vllm/v1/attention/backends/
|
|
- vllm/v1/attention/selector.py
|
|
- .buildkite/scripts/scheduled_integration_test/
|
|
- vllm/_aiter_ops.py
|
|
- vllm/platforms/rocm.py
|
|
commands:
|
|
- bash .buildkite/scripts/scheduled_integration_test/qwen30b_a3b_fp8_block_ep_eplb.sh 0.8 200 8020 2 1
|
|
|
|
- label: ":amd: (MI355) LM Eval Large Models FP8" # TBD
|
|
timeout_in_minutes: 45
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx950nightly, amdmi355]
|
|
dind: false
|
|
agent_pool: mi355_4
|
|
num_gpus: 4
|
|
optional: true
|
|
working_dir: "/vllm-workspace/.buildkite/lm-eval-harness"
|
|
source_file_dependencies:
|
|
- csrc/
|
|
- vllm/model_executor/layers/quantization
|
|
- vllm/model_executor/models/
|
|
- vllm/model_executor/model_loader/
|
|
- vllm/v1/attention/backends/
|
|
- vllm/v1/attention/selector.py
|
|
- vllm/_aiter_ops.py
|
|
- vllm/platforms/rocm.py
|
|
commands:
|
|
- export VLLM_USE_DEEP_GEMM=0
|
|
- pytest -s -v test_lm_eval_correctness.py --config-list-file=configs/models-large-rocm-tp4.txt --tp-size=4
|
|
|
|
- label: ":amd: (MI355) LM Eval Large Models" # TBD
|
|
timeout_in_minutes: 65
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx950nightly, amdmi355]
|
|
dind: false
|
|
agent_pool: mi355_8
|
|
optional: true
|
|
num_gpus: 8
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- vllm/model_executor/models/
|
|
- vllm/model_executor/model_loader/
|
|
- vllm/model_executor/layers/quantization/
|
|
- vllm/v1/attention/backends/
|
|
- vllm/v1/attention/selector.py
|
|
- vllm/model_executor/layers/layernorm.py
|
|
- csrc/
|
|
- vllm/_aiter_ops.py
|
|
- vllm/platforms/rocm.py
|
|
- tests/evals/
|
|
commands:
|
|
- export VLLM_WORKER_MULTIPROC_METHOD=spawn
|
|
- pytest -s -v evals/gsm8k/test_gsm8k_correctness.py --config-list-file=configs/models-gfx950-large.txt
|
|
|
|
- label: ":amd: (MI355) LM Eval DeepSeek R1 TP"
|
|
timeout_in_minutes: 65
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx950nightly, amdmi355]
|
|
dind: false
|
|
agent_pool: mi355_4
|
|
optional: true
|
|
num_gpus: 4
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- .buildkite/test_areas/lm_eval.yaml
|
|
- tests/evals/gsm8k/
|
|
- tests/utils.py
|
|
- tests/conftest.py
|
|
- vllm/config/
|
|
- vllm/distributed/
|
|
- vllm/model_executor/models/
|
|
- vllm/models/
|
|
- vllm/model_executor/model_loader/
|
|
- vllm/model_executor/layers/quantization/
|
|
- vllm/model_executor/layers/fused_moe/
|
|
- vllm/v1/attention/
|
|
- vllm/v1/spec_decode/
|
|
- vllm/v1/worker/gpu/spec_decode/
|
|
- vllm/v1/core/kv_cache_utils.py
|
|
- vllm/model_executor/layers/layernorm.py
|
|
- csrc/
|
|
- vllm/_aiter_ops.py
|
|
- vllm/platforms/rocm.py
|
|
- vllm/model_executor/layers/sparse_attn_indexer.py
|
|
commands:
|
|
- export VLLM_WORKER_MULTIPROC_METHOD=spawn
|
|
- pytest -s -v evals/gsm8k/test_gsm8k_correctness.py --config-list-file=configs/models-mi355-deepseek-r1-tp.txt
|
|
|
|
- label: ":amd: (MI355) LM Eval DeepSeek R1 DP"
|
|
timeout_in_minutes: 65
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx950nightly, amdmi355]
|
|
dind: false
|
|
agent_pool: mi355_4
|
|
optional: true
|
|
num_gpus: 4
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- .buildkite/test_areas/lm_eval.yaml
|
|
- tests/evals/gsm8k/
|
|
- tests/utils.py
|
|
- tests/conftest.py
|
|
- vllm/config/
|
|
- vllm/distributed/
|
|
- vllm/model_executor/models/
|
|
- vllm/models/
|
|
- vllm/model_executor/model_loader/
|
|
- vllm/model_executor/layers/quantization/
|
|
- vllm/model_executor/layers/fused_moe/
|
|
- vllm/v1/attention/
|
|
- vllm/v1/spec_decode/
|
|
- vllm/v1/worker/gpu/spec_decode/
|
|
- vllm/v1/core/kv_cache_utils.py
|
|
- vllm/model_executor/layers/layernorm.py
|
|
- csrc/
|
|
- vllm/_aiter_ops.py
|
|
- vllm/platforms/rocm.py
|
|
- vllm/model_executor/layers/sparse_attn_indexer.py
|
|
commands:
|
|
- export VLLM_WORKER_MULTIPROC_METHOD=spawn
|
|
- pytest -s -v evals/gsm8k/test_gsm8k_correctness.py --config-list-file=configs/models-mi355-deepseek-r1-dp.txt
|
|
|
|
- label: ":amd: (MI355) LM Eval DeepSeek V3-2 TP"
|
|
timeout_in_minutes: 65
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx950nightly, amdmi355]
|
|
dind: false
|
|
agent_pool: mi355_4
|
|
optional: true
|
|
num_gpus: 4
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- .buildkite/test_areas/lm_eval.yaml
|
|
- tests/evals/gsm8k/
|
|
- tests/utils.py
|
|
- tests/conftest.py
|
|
- vllm/config/
|
|
- vllm/distributed/
|
|
- vllm/model_executor/models/
|
|
- vllm/models/
|
|
- vllm/model_executor/model_loader/
|
|
- vllm/model_executor/layers/quantization/
|
|
- vllm/model_executor/layers/fused_moe/
|
|
- vllm/v1/attention/
|
|
- vllm/v1/spec_decode/
|
|
- vllm/v1/worker/gpu/spec_decode/
|
|
- vllm/v1/core/kv_cache_utils.py
|
|
- vllm/model_executor/layers/layernorm.py
|
|
- csrc/
|
|
- vllm/_aiter_ops.py
|
|
- vllm/platforms/rocm.py
|
|
- vllm/model_executor/layers/sparse_attn_indexer.py
|
|
commands:
|
|
- export VLLM_WORKER_MULTIPROC_METHOD=spawn
|
|
- pytest -s -v evals/gsm8k/test_gsm8k_correctness.py --config-list-file=configs/models-mi355-deepseek-v32-tp.txt
|
|
|
|
- label: ":amd: (MI355) LM Eval DeepSeek V3-2 DP"
|
|
timeout_in_minutes: 65
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx950nightly, amdmi355]
|
|
dind: false
|
|
agent_pool: mi355_4
|
|
optional: true
|
|
num_gpus: 4
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- .buildkite/test_areas/lm_eval.yaml
|
|
- tests/evals/gsm8k/
|
|
- tests/utils.py
|
|
- tests/conftest.py
|
|
- vllm/config/
|
|
- vllm/distributed/
|
|
- vllm/model_executor/models/
|
|
- vllm/models/
|
|
- vllm/model_executor/model_loader/
|
|
- vllm/model_executor/layers/quantization/
|
|
- vllm/model_executor/layers/fused_moe/
|
|
- vllm/v1/attention/
|
|
- vllm/v1/spec_decode/
|
|
- vllm/v1/worker/gpu/spec_decode/
|
|
- vllm/v1/core/kv_cache_utils.py
|
|
- vllm/model_executor/layers/layernorm.py
|
|
- csrc/
|
|
- vllm/_aiter_ops.py
|
|
- vllm/platforms/rocm.py
|
|
- vllm/model_executor/layers/sparse_attn_indexer.py
|
|
commands:
|
|
- export VLLM_WORKER_MULTIPROC_METHOD=spawn
|
|
- pytest -s -v evals/gsm8k/test_gsm8k_correctness.py --config-list-file=configs/models-mi355-deepseek-v32-dp.txt
|
|
|
|
- label: ":amd: (MI355) LM Eval Nemotron 3 Super"
|
|
timeout_in_minutes: 65
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx950nightly, amdmi355]
|
|
dind: false
|
|
agent_pool: mi355_4
|
|
optional: true
|
|
num_gpus: 4
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- .buildkite/test_areas/lm_eval.yaml
|
|
- tests/evals/gsm8k/
|
|
- tests/utils.py
|
|
- tests/conftest.py
|
|
- vllm/config/
|
|
- vllm/distributed/
|
|
- vllm/model_executor/models/
|
|
- vllm/models/
|
|
- vllm/model_executor/model_loader/
|
|
- vllm/model_executor/layers/quantization/
|
|
- vllm/model_executor/layers/fused_moe/
|
|
- vllm/v1/attention/
|
|
- vllm/v1/spec_decode/
|
|
- vllm/v1/worker/gpu/spec_decode/
|
|
- vllm/v1/core/kv_cache_utils.py
|
|
- vllm/model_executor/layers/layernorm.py
|
|
- csrc/
|
|
- vllm/_aiter_ops.py
|
|
- vllm/platforms/rocm.py
|
|
- vllm/model_executor/layers/mamba/
|
|
commands:
|
|
- export VLLM_WORKER_MULTIPROC_METHOD=spawn
|
|
- pytest -s -v evals/gsm8k/test_gsm8k_correctness.py --config-list-file=configs/models-mi355-nemotron-3-super.txt
|
|
|
|
#--------------------------------------------------------- mi355 · examples ----------------------------------------------------------#
|
|
|
|
- label: ":amd: (MI355 DPX) Examples" # TBD
|
|
timeout_in_minutes: 75
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx950nightly, amdmi355]
|
|
dind: false
|
|
agent_pool: mi355_dpx
|
|
optional: true
|
|
working_dir: "/vllm-workspace/examples"
|
|
source_file_dependencies:
|
|
- vllm/entrypoints
|
|
- vllm/multimodal
|
|
- examples/
|
|
- vllm/platforms/rocm.py
|
|
commands:
|
|
# Basic
|
|
# Multi-modal models
|
|
# Pooling models
|
|
# Features demo
|
|
- pip install --no-deps tensorizer
|
|
- python3 basic/offline_inference/chat.py
|
|
- python3 basic/offline_inference/generate.py --model facebook/opt-125m
|
|
- python3 basic/offline_inference/generate.py --model meta-llama/Llama-2-13b-chat-hf --cpu-offload-gb 10
|
|
- python3 basic/offline_inference/classify.py
|
|
- python3 basic/offline_inference/embed.py
|
|
- python3 basic/offline_inference/score.py
|
|
- python3 generate/multimodal/audio_language_offline.py --seed 0
|
|
- python3 generate/multimodal/vision_language_offline.py --seed 0
|
|
- python3 generate/multimodal/vision_language_multi_image_offline.py --seed 0
|
|
- python3 generate/multimodal/encoder_decoder_multimodal_offline.py --model-type whisper --seed 0
|
|
- python3 pooling/embed/vision_embedding_offline.py --seed 0
|
|
- python3 features/automatic_prefix_caching/prefix_caching_offline.py
|
|
- python3 deployment/llm_engine_example.py
|
|
- python3 features/tensorize_vllm_model.py --model facebook/opt-125m serialize --serialized-directory /tmp/ --suffix v1 && python3 features/tensorize_vllm_model.py --model facebook/opt-125m deserialize --path-to-tensors /tmp/vllm/facebook/opt-125m/v1/model.tensors
|
|
- python3 features/speculative_decoding/spec_decode_offline.py --test --method eagle --num_spec_tokens 3 --dataset-name hf --dataset-path philschmid/mt-bench --num-prompts 80 --temp 0 --top-p 1.0 --top-k -1 --tp 1 --enable-chunked-prefill --max-model-len 2048
|
|
- python3 features/speculative_decoding/spec_decode_offline.py --test --method eagle3 --num_spec_tokens 3 --dataset-name hf --dataset-path philschmid/mt-bench --num-prompts 80 --temp 0 --top-p 1.0 --top-k -1 --tp 1 --enable-chunked-prefill --max-model-len 1536
|
|
|
|
#---------------------------------------------------------- mi355 · kernels ----------------------------------------------------------#
|
|
|
|
- label: ":amd: (MI355) mHC Kernels"
|
|
timeout_in_minutes: 50
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx950nightly, amdmi355]
|
|
dind: false
|
|
agent_pool: mi355_1
|
|
optional: true
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- csrc/
|
|
- vllm/
|
|
- tests/kernels/
|
|
- vllm/platforms/rocm.py
|
|
commands:
|
|
- pytest -v -s kernels/mhc
|
|
|
|
- label: ":amd: (MI355) FLA Ops Kernels"
|
|
timeout_in_minutes: 50
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx950nightly, amdmi355]
|
|
dind: false
|
|
agent_pool: mi355_1
|
|
optional: true
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- csrc/
|
|
- vllm/
|
|
- tests/kernels/
|
|
- vllm/platforms/rocm.py
|
|
commands:
|
|
- pytest -v -s kernels/fla
|
|
|
|
- label: ":amd: (MI355) Miscellaneous Kernels"
|
|
timeout_in_minutes: 55
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx950nightly, amdmi355]
|
|
dind: false
|
|
agent_pool: mi355_1
|
|
optional: true
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- csrc/
|
|
- vllm/
|
|
- tests/kernels/
|
|
- vllm/platforms/rocm.py
|
|
commands:
|
|
- pytest -v -s
|
|
kernels/test_apply_repetition_penalties.py
|
|
kernels/test_awq_int4_to_int8.py
|
|
kernels/test_bf16_skinny_gemm.py
|
|
kernels/test_cache_kernels.py
|
|
kernels/test_compressor_kv_cache.py
|
|
kernels/test_cp_gather_fp8.py
|
|
kernels/test_deepseek_v4_cpu_kernels.py
|
|
kernels/test_dsv41_mega_attn_layouts.py
|
|
kernels/test_engram.py
|
|
kernels/test_flashinfer_bf16_gemm.py
|
|
kernels/test_flex_attention.py
|
|
kernels/test_fp32_router_gemm.py
|
|
kernels/test_fused_deepseek_v32_norm_rope.py
|
|
kernels/test_fused_indexer_q_rope_quant.py
|
|
kernels/test_fused_inv_rope_fp8_quant.py
|
|
kernels/test_fused_minimax_m3_qknorm_rope_kv_insert.py
|
|
kernels/test_fused_quant_activation.py
|
|
kernels/test_gate_linear_rocm_dispatch.py
|
|
kernels/test_gemm_rs_ar.py
|
|
kernels/test_kpool_decode_update_batched.py
|
|
kernels/test_minimax_m3_amd_ops.py
|
|
kernels/test_minimax_m3_sparse_attn_fp8_scale.py
|
|
kernels/test_onednn.py
|
|
kernels/test_relu2_fp8_quant.py
|
|
kernels/test_rocm_fp32_router_gemm.py
|
|
--deselect=tests/kernels/test_bf16_skinny_gemm.py::test_every_dsv3_routed_shape_is_instantiated
|
|
|
|
- label: ":amd: (MI355 DPX) Attention DiffKV Kernels"
|
|
timeout_in_minutes: 35
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx950nightly, amdmi355]
|
|
dind: false
|
|
agent_pool: mi355_dpx
|
|
num_gpus: 1
|
|
optional: true
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- vllm/v1/attention/ops/triton_unified_attention_diffkv.py
|
|
- vllm/v1/attention/backends/triton_attn_diffkv.py
|
|
- tests/kernels/attention/test_triton_unified_attention.py
|
|
- tests/kernels/attention/test_triton_unified_attention_diffkv.py
|
|
- vllm/platforms/rocm.py
|
|
commands:
|
|
- pytest -v -s kernels/attention/test_triton_unified_attention_diffkv.py
|
|
|
|
- label: ":amd: (MI355) MiniMax Reduce RMS Kernels" # TBD
|
|
timeout_in_minutes: 45
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx950nightly, amdmi355]
|
|
dind: false
|
|
agent_pool: mi355_2
|
|
num_gpus: 2
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- vllm/model_executor/layers/minimax_rms_norm/
|
|
- vllm/distributed/
|
|
- vllm/_aiter_ops.py
|
|
- vllm/envs.py
|
|
- vllm/platforms/rocm.py
|
|
- tests/kernels/core/test_minimax_reduce_rms.py
|
|
- tests/utils.py
|
|
- csrc/libtorch_stable/minimax_reduce_rms_kernel.cu
|
|
- csrc/libtorch_stable/minimax_reduce_rms_kernel.h
|
|
- csrc/libtorch_stable/ops.h
|
|
- csrc/libtorch_stable/torch_bindings.cpp
|
|
- vllm/model_executor/layers/mamba/linear/minimax_linear_attn.py
|
|
optional: true
|
|
commands:
|
|
- pytest -v -s kernels/core/test_minimax_reduce_rms.py
|
|
|
|
- label: ":amd: (MI355 DPX) Kernels" # TBD
|
|
timeout_in_minutes: 25
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx950nightly, amdmi355]
|
|
dind: false
|
|
agent_pool: mi355_dpx
|
|
working_dir: "/vllm-workspace/"
|
|
source_file_dependencies:
|
|
- csrc/quantization/fp4/
|
|
- csrc/attention/mla/
|
|
- csrc/quantization/cutlass_w8a8/moe/
|
|
- vllm/model_executor/layers/fused_moe/cutlass_moe.py
|
|
- vllm/v1/attention/backends/triton_attn.py
|
|
- vllm/v1/attention/backends/rocm_attn.py
|
|
- vllm/v1/attention/backends/rocm_aiter_fa.py
|
|
- vllm/v1/attention/backends/rocm_aiter_unified_attn.py
|
|
- vllm/v1/attention/backends/mla/aiter_triton_mla.py
|
|
- vllm/v1/attention/backends/mla/rocm_aiter_mla.py
|
|
- vllm/v1/attention/selector.py
|
|
- vllm/platforms/rocm.py
|
|
- vllm/_aiter_ops.py
|
|
- examples/basic/offline_inference/chat.py
|
|
- tests/kernels/attention/test_attention_selector.py
|
|
optional: true
|
|
commands:
|
|
- amd-smi
|
|
- python3 examples/basic/offline_inference/chat.py --attention-backend TRITON_ATTN
|
|
- pytest -v -s tests/kernels/attention/test_attention_selector.py
|
|
|
|
- label: ":amd: (MI355 DPX) MLA Kernels"
|
|
timeout_in_minutes: 40
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx950nightly, amdmi355]
|
|
dind: false
|
|
agent_pool: mi355_dpx
|
|
num_gpus: 1
|
|
optional: true
|
|
working_dir: "/vllm-workspace/"
|
|
source_file_dependencies:
|
|
- vllm/v1/attention/backends/mla/rocm_aiter_mla.py
|
|
- vllm/v1/attention/selector.py
|
|
- vllm/platforms/rocm.py
|
|
- vllm/_aiter_ops.py
|
|
- tests/kernels/attention/test_rocm_aiter_mla_decode.py
|
|
- tests/kernels/attention/test_rocm_aiter_mla_decode_metadata.py
|
|
- tests/kernels/attention/test_rocm_aiter_mla_fp8_support.py
|
|
- tests/kernels/attention/test_rocm_aiter_mla_op_registration.py
|
|
commands:
|
|
- amd-smi
|
|
- pytest -v -s tests/kernels/attention/test_rocm_aiter_mla_decode_metadata.py
|
|
- pytest -v -s tests/kernels/attention/test_rocm_aiter_mla_decode.py
|
|
- pytest -v -s tests/kernels/attention/test_rocm_aiter_mla_fp8_support.py
|
|
- pytest -v -s tests/kernels/attention/test_rocm_aiter_mla_op_registration.py
|
|
|
|
- label: ":amd: (MI355 DPX) DeepSeek V4 Fused Kernels"
|
|
timeout_in_minutes: 60
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx950nightly, amdmi355]
|
|
dind: false
|
|
agent_pool: mi355_dpx
|
|
num_gpus: 1
|
|
optional: true
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- csrc/libtorch_stable/fused_deepseek_v4_qnorm_rope_kv_insert_kernel.cu
|
|
- vllm/models/deepseek_v4/common/ops/
|
|
- tests/kernels/test_fused_deepseek_v4_qnorm_rope_kv_insert.py
|
|
- vllm/platforms/rocm.py
|
|
commands:
|
|
- pytest -v -s kernels/test_fused_deepseek_v4_qnorm_rope_kv_insert.py
|
|
|
|
- label: ":amd: (MI355 DPX) Attention Kernels Shard %N" # TBD
|
|
timeout_in_minutes: 90
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx950nightly, amdmi355]
|
|
dind: false
|
|
agent_pool: mi355_dpx
|
|
parallelism: 2
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- csrc/attention/
|
|
- vllm/v1/attention
|
|
- vllm/model_executor/layers/attention
|
|
- tests/kernels/attention
|
|
- vllm/_aiter_ops.py
|
|
- vllm/envs.py
|
|
- vllm/platforms/rocm.py
|
|
- vllm/utils/flashinfer.py
|
|
optional: true
|
|
commands:
|
|
- pytest -v -s kernels/attention --ignore=kernels/attention/test_triton_unified_attention_diffkv.py --ignore=kernels/attention/test_rocm_aiter_mla_decode.py --ignore=kernels/attention/test_rocm_aiter_mla_decode_metadata.py --ignore=kernels/attention/test_rocm_aiter_mla_fp8_support.py --ignore=kernels/attention/test_rocm_aiter_mla_op_registration.py --shard-id=$$BUILDKITE_PARALLEL_JOB --num-shards=$$BUILDKITE_PARALLEL_JOB_COUNT
|
|
|
|
- label: ":amd: (MI355 DPX) MoE Kernels Shard %N" # TBD
|
|
timeout_in_minutes: 55
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx950nightly, amdmi355]
|
|
dind: false
|
|
agent_pool: mi355_dpx
|
|
parallelism: 5
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- csrc/quantization/cutlass_w8a8/moe/
|
|
- csrc/moe/
|
|
- csrc/rocm/
|
|
- tests/kernels/moe
|
|
- tests/kernels/utils.py
|
|
- vllm/model_executor/layers/fused_moe/
|
|
- vllm/model_executor/layers/quantization/mxfp4.py
|
|
- vllm/model_executor/layers/quantization/utils/quant_utils.py
|
|
- vllm/distributed/device_communicators/
|
|
- vllm/envs.py
|
|
- vllm/config
|
|
- vllm/_aiter_ops.py
|
|
- vllm/platforms/rocm.py
|
|
optional: true
|
|
commands:
|
|
- pytest -v -s kernels/moe
|
|
--ignore=kernels/moe/test_modular_oai_triton_moe.py
|
|
--ignore=kernels/moe/test_gpt_oss_triton_kernels.py
|
|
--ignore=kernels/moe/test_moe.py
|
|
--ignore=kernels/moe/test_block_int8.py
|
|
--ignore=kernels/moe/test_triton_moe_no_act_mul.py
|
|
--ignore=kernels/moe/test_triton_moe_ptpc_fp8.py
|
|
--ignore=kernels/moe/test_deepep_moe.py
|
|
--ignore=kernels/moe/test_moe_layer.py
|
|
--shard-id=$$BUILDKITE_PARALLEL_JOB --num-shards=$$BUILDKITE_PARALLEL_JOB_COUNT
|
|
- pytest -v -s kernels/moe/test_modular_oai_triton_moe.py --shard-id=$$BUILDKITE_PARALLEL_JOB --num-shards=$$BUILDKITE_PARALLEL_JOB_COUNT
|
|
|
|
- label: ":amd: (MI355) FusedMoE Layer Kernels" # TBD
|
|
timeout_in_minutes: 35
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx950nightly, amdmi355]
|
|
dind: false
|
|
agent_pool: mi355_2
|
|
num_gpus: 2
|
|
optional: true
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- csrc/moe/
|
|
- csrc/rocm/
|
|
- tests/kernels/moe
|
|
- vllm/model_executor/layers/fused_moe/
|
|
- vllm/model_executor/layers/quantization/
|
|
- vllm/distributed/
|
|
- vllm/config/
|
|
- vllm/forward_context.py
|
|
- vllm/v1/worker/workspace.py
|
|
- vllm/utils/import_utils.py
|
|
- vllm/utils/math_utils.py
|
|
- vllm/utils/torch_utils.py
|
|
- vllm/platforms/
|
|
- vllm/_aiter_ops.py
|
|
- csrc/quantization/cutlass_w8a8/moe/
|
|
- vllm/distributed/device_communicators/
|
|
- vllm/config
|
|
commands:
|
|
# MoRI graph cases run in the Fault Tolerance group's healthy EP baseline.
|
|
- pytest -v -s kernels/moe/test_moe_layer.py -k "not test_moe_layer_mori_graph"
|
|
|
|
- label: ":amd: (MI355) MoRI EP Numerics"
|
|
timeout_in_minutes: 20
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx950nightly, amdmi355]
|
|
dind: false
|
|
agent_pool: mi355_2
|
|
num_gpus: 2
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- tests/kernels/moe/test_mori_moe.py
|
|
- tests/kernels/moe/utils.py
|
|
- tests/kernels/moe/modular_kernel_tools/parallel_utils.py
|
|
- tests/kernels/utils.py
|
|
- tests/utils.py
|
|
- tests/conftest.py
|
|
- vllm/utils/torch_utils.py
|
|
- vllm/model_executor/layers/fused_moe/
|
|
- vllm/model_executor/layers/quantization/
|
|
- vllm/distributed/
|
|
- vllm/config/
|
|
- vllm/forward_context.py
|
|
- vllm/v1/worker/workspace.py
|
|
- vllm/platforms/
|
|
- vllm/_aiter_ops.py
|
|
- requirements/rocm-test.txt
|
|
optional: true
|
|
commands:
|
|
- amd-smi
|
|
- VLLM_ROCM_USE_AITER=1 VLLM_ROCM_USE_AITER_MOE=1 VLLM_ROCM_USE_AITER_FUSION_SHARED_EXPERTS=0 pytest -v -s kernels/moe/test_mori_moe.py
|
|
|
|
- label: ":amd: (MI355) Quantization Kernels Shard %N"
|
|
timeout_in_minutes: 120
|
|
parallelism: 6
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx950nightly, amdmi355]
|
|
dind: false
|
|
agent_pool: mi355_1
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- csrc/quantization/
|
|
- csrc/rocm/
|
|
- vllm/model_executor/layers/quantization
|
|
- tests/kernels/quantization
|
|
- tests/kernels/quant_utils.py
|
|
- tests/kernels/utils.py
|
|
- vllm/_aiter_ops.py
|
|
- vllm/kernels/aiter_ops.py
|
|
- vllm/_custom_ops.py
|
|
- vllm/envs.py
|
|
- vllm/platforms/rocm.py
|
|
- vllm/model_executor/kernels/
|
|
- vllm/v1/attention/backends/rocm_aiter_fa.py
|
|
- vllm/config/
|
|
- tests/kernels/quantization/test_rocm_skinny_gemms.py
|
|
optional: true
|
|
commands:
|
|
- pytest -v -s kernels/quantization -k "not test_nvfp4_dense_emulation" --shard-id=$$BUILDKITE_PARALLEL_JOB --num-shards=$$BUILDKITE_PARALLEL_JOB_COUNT
|
|
|
|
- label: ":amd: (MI355) NVFP4 Dense Emulation Shard %N"
|
|
timeout_in_minutes: 10
|
|
parallelism: 3
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx950nightly, amdmi355]
|
|
dind: false
|
|
agent_pool: mi355_1
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- tests/kernels/quantization/test_nvfp4_emulation.py
|
|
- tests/kernels/quantization/nvfp4_utils.py
|
|
- tests/conftest.py
|
|
- vllm/model_executor/kernels/linear/
|
|
- vllm/model_executor/layers/quantization/utils/nvfp4_emulation_utils.py
|
|
- vllm/config/kernel.py
|
|
- vllm/platforms/
|
|
- vllm/envs.py
|
|
- vllm/utils/torch_utils.py
|
|
- requirements/rocm-test.txt
|
|
optional: true
|
|
commands:
|
|
- amd-smi
|
|
- pytest -v -s kernels/quantization/test_nvfp4_emulation.py -k test_nvfp4_dense_emulation --shard-id=$$BUILDKITE_PARALLEL_JOB --num-shards=$$BUILDKITE_PARALLEL_JOB_COUNT
|
|
|
|
- label: ":amd: (MI355 DPX) FP8 MoE Kernels" # TBD
|
|
timeout_in_minutes: 40
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx950nightly, amdmi355]
|
|
dind: false
|
|
agent_pool: mi355_dpx
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- csrc/moe/
|
|
- vllm/model_executor/layers/fused_moe/
|
|
- tests/kernels/moe/test_deepep_moe.py
|
|
- vllm/_aiter_ops.py
|
|
- vllm/platforms/rocm.py
|
|
- vllm/envs.py
|
|
- tests/kernels/moe/test_gpt_oss_triton_kernels.py
|
|
- tests/kernels/moe/test_moe.py
|
|
- tests/kernels/moe/test_block_int8.py
|
|
- tests/kernels/moe/test_triton_moe_no_act_mul.py
|
|
- tests/kernels/moe/test_triton_moe_ptpc_fp8.py
|
|
optional: true
|
|
commands:
|
|
- pytest -v -s kernels/moe/test_gpt_oss_triton_kernels.py
|
|
- pytest -v -s kernels/moe/test_moe.py
|
|
- pytest -v -s kernels/moe/test_block_int8.py
|
|
- pytest -v -s kernels/moe/test_triton_moe_no_act_mul.py
|
|
- pytest -v -s kernels/moe/test_triton_moe_ptpc_fp8.py
|
|
|
|
- label: ":amd: (MI355) DeepEP FP8 MoE Kernels" # TBD
|
|
timeout_in_minutes: 55
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx950nightly, amdmi355]
|
|
dind: false
|
|
agent_pool: mi355_2
|
|
num_gpus: 2
|
|
optional: true
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- csrc/moe/
|
|
- csrc/quantization/w8a8/cutlass/moe/
|
|
- vllm/model_executor/layers/fused_moe/
|
|
- tests/kernels/moe/test_deepep_moe.py
|
|
- vllm/_aiter_ops.py
|
|
- vllm/platforms/rocm.py
|
|
- vllm/envs.py
|
|
commands:
|
|
- pytest -v -s kernels/moe/test_deepep_moe.py
|
|
|
|
#----------------------------------------------------------- mi355 · lora ------------------------------------------------------------#
|
|
|
|
- label: ":amd: (MI355 DPX) LoRA Shard %N" # TBD
|
|
timeout_in_minutes: 85
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx950nightly, amdmi355]
|
|
dind: false
|
|
agent_pool: mi355_dpx
|
|
parallelism: 4
|
|
optional: true
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- vllm/lora
|
|
- tests/lora
|
|
- vllm/platforms/rocm.py
|
|
commands:
|
|
- pytest -v -s lora --shard-id=$$BUILDKITE_PARALLEL_JOB --num-shards=$$BUILDKITE_PARALLEL_JOB_COUNT --ignore=lora/test_chatglm3_tp.py --ignore=lora/test_llama_tp.py --ignore=lora/test_qwen3_with_multi_loras.py --ignore=lora/test_olmoe_tp.py --ignore=lora/test_deepseekv2_tp.py --ignore=lora/test_gptoss_tp.py --ignore=lora/test_qwen3moe_tp.py --ignore=lora/test_qwen35_densemodel_lora.py
|
|
|
|
#------------------------------------------------------ mi355 · models / basic -------------------------------------------------------#
|
|
|
|
- label: ":amd: (MI355) Basic Models (Other)" # TBD
|
|
timeout_in_minutes: 65
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx950nightly, amdmi355]
|
|
dind: false
|
|
agent_pool: mi355_1
|
|
optional: true
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- vllm/
|
|
- tests/models/test_terratorch.py
|
|
- tests/models/transformers/test_backend.py
|
|
- tests/models/test_registry.py
|
|
- tests/models/test_hyv4_rocm.py
|
|
- tests/models/test_vision.py
|
|
- tests/models/test_deepseek_v4_vl_rocm.py
|
|
commands:
|
|
- pytest -v -s models/test_terratorch.py models/transformers/test_backend.py models/test_registry.py models/test_deepseek_v4_vl_rocm.py
|
|
- VLLM_ROCM_USE_AITER=1 pytest -v -s models/test_hyv4_rocm.py -m 'not distributed'
|
|
- pytest -v -s models/test_vision.py::test_simple_mrope_vision_model_spatial_merge
|
|
|
|
- label: ":amd: (MI355 DPX) GLM5Next"
|
|
timeout_in_minutes: 35
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx950nightly, amdmi355]
|
|
dind: false
|
|
agent_pool: mi355_dpx
|
|
num_gpus: 1
|
|
optional: true
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- vllm/models/glm5next/
|
|
- tests/models/glm5next/
|
|
- vllm/platforms/rocm.py
|
|
commands:
|
|
- pytest -v -s models/glm5next
|
|
|
|
- label: ":amd: (MI355 DPX) Basic Models (Extra Initialization) Shard %N" # TBD
|
|
timeout_in_minutes: 60
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx950nightly, amdmi355]
|
|
dind: false
|
|
agent_pool: mi355_dpx
|
|
parallelism: 6
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- vllm/model_executor/models/
|
|
- vllm/model_executor/layers/
|
|
- tests/models/test_initialization.py
|
|
- tests/models/registry.py
|
|
- vllm/_aiter_ops.py
|
|
- vllm/platforms/rocm.py
|
|
optional: true
|
|
commands:
|
|
- pytest -v -s models/test_initialization.py -k 'not test_can_initialize_small_subset' --num-shards=$$BUILDKITE_PARALLEL_JOB_COUNT --shard-id=$$BUILDKITE_PARALLEL_JOB
|
|
|
|
- label: ":amd: (MI355) Inkling" # TBD
|
|
timeout_in_minutes: 65
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx950nightly, amdmi355]
|
|
dind: false
|
|
agent_pool: mi355_1
|
|
optional: true
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- vllm/models/inkling/
|
|
- vllm/cute_utils/
|
|
- cmake/external_projects/tml_fa4.cmake
|
|
- tests/models/inkling/
|
|
- vllm/platforms/rocm.py
|
|
commands:
|
|
- pytest -v -s models/inkling/rocm
|
|
|
|
#----------------------------------------------------- mi355 · models / language -----------------------------------------------------#
|
|
|
|
- label: ":amd: (MI355 DPX) Language Models (Standard)" # TBD
|
|
timeout_in_minutes: 60
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx950nightly, amdmi355]
|
|
dind: false
|
|
agent_pool: mi355_dpx
|
|
optional: true
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- vllm/
|
|
- tests/models/language
|
|
commands:
|
|
- pip freeze | grep -E 'torch'
|
|
- pytest -v -s models/language -m 'core_model and (not slow_test)'
|
|
|
|
- label: ":amd: (MI355 DPX) Language Models (Extra Standard) Shard %N" # TBD
|
|
timeout_in_minutes: 40
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx950nightly, amdmi355]
|
|
dind: false
|
|
agent_pool: mi355_dpx
|
|
optional: true
|
|
parallelism: 2
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- vllm/model_executor/models/
|
|
- vllm/model_executor/model_loader/
|
|
- vllm/model_executor/layers/
|
|
- vllm/v1/attention/backends/
|
|
- vllm/v1/attention/selector.py
|
|
- tests/models/language/pooling/test_embedding.py
|
|
- tests/models/language/generation/test_common.py
|
|
- tests/models/language/pooling/test_classification.py
|
|
- vllm/_aiter_ops.py
|
|
- vllm/platforms/rocm.py
|
|
commands:
|
|
- pip freeze | grep -E 'torch'
|
|
- pytest -v -s models/language -m 'core_model and slow_test' --num-shards=$$BUILDKITE_PARALLEL_JOB_COUNT --shard-id=$$BUILDKITE_PARALLEL_JOB
|
|
|
|
- label: ":amd: (MI355 DPX) Language Models (Hybrid) Shard %N" # TBD
|
|
timeout_in_minutes: 65
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx950nightly, amdmi355]
|
|
dind: false
|
|
agent_pool: mi355_dpx
|
|
parallelism: 2
|
|
optional: true
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- vllm/
|
|
- tests/models/language/generation
|
|
commands:
|
|
- export GIT_CONFIG_COUNT=1
|
|
- export GIT_CONFIG_KEY_0=http.version
|
|
- export GIT_CONFIG_VALUE_0=HTTP/1.1
|
|
- MAMBA_FORCE_BUILD=TRUE uv pip install --system --no-build-isolation 'git+https://github.com/AndreasKaratzas/mamba@fix-rocm-7.0-warp-size-constexpr'
|
|
- CAUSAL_CONV1D_FORCE_BUILD=TRUE uv pip install --system --no-build-isolation 'git+https://github.com/Dao-AILab/causal-conv1d@v1.6.0'
|
|
- pytest -v -s models/language/generation -m hybrid_model -k 'not granite-4.0-tiny-preview' --num-shards=$$BUILDKITE_PARALLEL_JOB_COUNT --shard-id=$$BUILDKITE_PARALLEL_JOB
|
|
|
|
- label: ":amd: (MI355 DPX) Granite Language Model Compatibility"
|
|
timeout_in_minutes: 80
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx950nightly, amdmi355]
|
|
dind: false
|
|
agent_pool: mi355_dpx
|
|
num_gpus: 1
|
|
optional: true
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- vllm/
|
|
- "!vllm/distributed/kv_transfer/"
|
|
- tests/models/language/generation
|
|
commands:
|
|
- export GIT_CONFIG_COUNT=1
|
|
- export GIT_CONFIG_KEY_0=http.version
|
|
- export GIT_CONFIG_VALUE_0=HTTP/1.1
|
|
- MAMBA_FORCE_BUILD=TRUE uv pip install --system --no-build-isolation 'git+https://github.com/AndreasKaratzas/mamba@fix-rocm-7.0-warp-size-constexpr'
|
|
- CAUSAL_CONV1D_FORCE_BUILD=TRUE 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 hybrid_model -k 'granite-4.0-tiny-preview'
|
|
|
|
- label: ":amd: (MI355 DPX) Language Models (Extended Generation)" # TBD
|
|
timeout_in_minutes: 95
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx950nightly, amdmi355]
|
|
dind: false
|
|
agent_pool: mi355_dpx
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- vllm/
|
|
- tests/models/language/generation
|
|
optional: true
|
|
commands:
|
|
- MAMBA_FORCE_BUILD=TRUE uv pip install --system --no-build-isolation 'git+https://github.com/AndreasKaratzas/mamba@rocm-7.0-v2.3.0'
|
|
- CAUSAL_CONV1D_FORCE_BUILD=TRUE 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)'
|
|
|
|
- label: ":amd: (MI355 DPX) Language Models (Extended Pooling) Shard %N"
|
|
timeout_in_minutes: 95
|
|
parallelism: 4
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx950nightly, amdmi355]
|
|
dind: false
|
|
agent_pool: mi355_dpx
|
|
optional: true
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- vllm/
|
|
- tests/models/language/pooling
|
|
commands:
|
|
- pytest -v -s models/language/pooling -m 'not core_model' --num-shards=$$BUILDKITE_PARALLEL_JOB_COUNT --shard-id=$$BUILDKITE_PARALLEL_JOB
|
|
|
|
- label: ":amd: (MI355 DPX) Language Models (PPL)" # TBD
|
|
timeout_in_minutes: 65
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx950nightly, amdmi355]
|
|
dind: false
|
|
agent_pool: mi355_dpx
|
|
optional: true
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- vllm/model_executor/models/qwen3_5.py
|
|
- vllm/model_executor/models/qwen3_5_mtp.py
|
|
- vllm/transformers_utils/configs/qwen3_5.py
|
|
- vllm/transformers_utils/configs/qwen3_5_moe.py
|
|
- vllm/model_executor/models/qwen2.py
|
|
- vllm/model_executor/models/qwen3.py
|
|
- vllm/model_executor/models/qwen3_next.py
|
|
- vllm/model_executor/models/qwen3_next_mtp.py
|
|
- vllm/third_party/flash_linear_attention/ops/
|
|
- vllm/_aiter_ops.py
|
|
- vllm/v1/attention/backends/triton_attn.py
|
|
- vllm/v1/attention/backends/rocm_attn.py
|
|
- vllm/v1/attention/backends/rocm_aiter_unified_attn.py
|
|
- vllm/v1/attention/backends/rocm_aiter_fa.py
|
|
- vllm/v1/attention/backends/flex_attention.py
|
|
- vllm/v1/attention/ops/
|
|
- vllm/platforms/rocm.py
|
|
- tests/models/language/generation_ppl_test
|
|
- vllm/model_executor/layers/fla/ops/
|
|
- vllm/
|
|
commands:
|
|
- pytest -v -s models/language/generation_ppl_test
|
|
|
|
#---------------------------------------------------- mi355 · models / multimodal ----------------------------------------------------#
|
|
|
|
- label: ":amd: (MI355 DPX) Multimodal Models (Standard) 1: qwen2" # TBD
|
|
timeout_in_minutes: 65
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx950nightly, amdmi355]
|
|
dind: false
|
|
agent_pool: mi355_dpx
|
|
optional: true
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- vllm/
|
|
- tests/models/multimodal
|
|
commands:
|
|
- pytest -v -s models/multimodal/generation/test_common.py -m core_model -k "qwen2"
|
|
- pytest -v -s models/multimodal/generation/test_ultravox.py -m core_model
|
|
|
|
- label: ":amd: (MI355 DPX) Multimodal Models (Standard) 2: qwen3 + gemma" # TBD
|
|
timeout_in_minutes: 55
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx950nightly, amdmi355]
|
|
dind: false
|
|
agent_pool: mi355_dpx
|
|
optional: true
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- vllm/
|
|
- tests/models/multimodal
|
|
commands:
|
|
- pytest -v -s models/multimodal/generation/test_common.py -m core_model -k "qwen3 or gemma"
|
|
- pytest -v -s models/multimodal/generation/test_mm_prefix_lm.py -m core_model
|
|
- pytest -v -s models/multimodal/generation/test_qwen2_5_vl.py -m core_model
|
|
|
|
- label: ":amd: (MI355 DPX) Multimodal Models (Extended Generation 1)" # TBD
|
|
timeout_in_minutes: 90
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx950nightly, amdmi355]
|
|
dind: false
|
|
agent_pool: mi355_dpx
|
|
optional: true
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- vllm/
|
|
- tests/models/multimodal/generation
|
|
- tests/models/multimodal/test_mapping.py
|
|
commands:
|
|
- pytest -v -s models/multimodal/generation -m 'not core_model' --ignore models/multimodal/generation/test_common.py
|
|
- pytest -v -s models/multimodal/test_mapping.py
|
|
|
|
- label: ":amd: (MI355 DPX) Multimodal Models (Extended Generation 3)" # TBD
|
|
timeout_in_minutes: 90
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx950nightly, amdmi355]
|
|
dind: false
|
|
agent_pool: mi355_dpx
|
|
optional: true
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- vllm/
|
|
- tests/models/multimodal/generation
|
|
commands:
|
|
- pytest -v -s models/multimodal/generation/test_common.py -m 'split(group=1) and not core_model'
|
|
|
|
- label: ":amd: (MI355 DPX) Multimodal Models (Extended Pooling)" # TBD
|
|
timeout_in_minutes: 75
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx950nightly, amdmi355]
|
|
dind: false
|
|
agent_pool: mi355_dpx
|
|
optional: true
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- vllm/
|
|
- tests/models/multimodal/pooling
|
|
commands:
|
|
- pytest -v -s models/multimodal/pooling -m 'not core_model'
|
|
|
|
- label: ":amd: (MI355 DPX) Multimodal Models (Standard) 4: other + whisper" # TBD
|
|
timeout_in_minutes: 70
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx950nightly, amdmi355]
|
|
dind: false
|
|
agent_pool: mi355_dpx
|
|
optional: true
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- vllm/
|
|
- tests/models/multimodal
|
|
commands:
|
|
- pytest -v -s models/multimodal -m core_model --ignore models/multimodal/generation/test_common.py --ignore models/multimodal/generation/test_ultravox.py --ignore models/multimodal/generation/test_qwen2_5_vl.py --ignore models/multimodal/generation/test_qwen2_vl.py --ignore models/multimodal/generation/test_whisper.py --ignore models/multimodal/generation/test_mm_prefix_lm.py --ignore models/multimodal/generation/test_memory_leak.py --ignore models/multimodal/generation/test_vit_cudagraph.py --ignore models/multimodal/processing
|
|
- pytest -v -s models/multimodal/generation/test_vit_cudagraph.py -m core_model
|
|
- pytest models/multimodal/generation/test_memory_leak.py -m core_model
|
|
- cd .. && VLLM_WORKER_MULTIPROC_METHOD=spawn pytest -v -s tests/models/multimodal/generation/test_whisper.py -m core_model
|
|
|
|
#----------------------------------------------------- mi355 · models / quantized -----------------------------------------------------#
|
|
|
|
- label: ":amd: (MI355 DPX) Quantized Models" # TBD
|
|
timeout_in_minutes: 80
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx950nightly, amdmi355]
|
|
dind: false
|
|
agent_pool: mi355_dpx
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- vllm/model_executor/layers/quantization
|
|
- tests/models/quantization
|
|
- vllm/_aiter_ops.py
|
|
- vllm/platforms/rocm.py
|
|
- vllm/model_executor/model_loader/
|
|
optional: true
|
|
commands:
|
|
- unset VLLM_USE_V2_MODEL_RUNNER
|
|
- pytest -v -s models/quantization
|
|
|
|
#------------------------------------------------------- mi355 · quantization --------------------------------------------------------#
|
|
|
|
- label: ":amd: (MI355 DPX) Quantization Shard %N"
|
|
timeout_in_minutes: 115
|
|
parallelism: 4
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx950nightly, amdmi355]
|
|
dind: false
|
|
agent_pool: mi355_dpx
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- csrc/
|
|
- vllm/model_executor/layers/quantization
|
|
- tests/quantization
|
|
- vllm/_aiter_ops.py
|
|
- vllm/platforms/rocm.py
|
|
- vllm/model_executor/layers/fused_moe/oracle/unquantized.py
|
|
- vllm/model_executor/layers/fused_moe/unquantized_fused_moe_method.py
|
|
- tests/rocm/test_moe_weight_replay.py
|
|
optional: true
|
|
commands:
|
|
- unset VLLM_USE_V2_MODEL_RUNNER
|
|
- uv pip install --system torchao==0.17.0
|
|
- uv pip install --system conch-triton-kernels
|
|
- VLLM_TEST_FORCE_LOAD_FORMAT=auto pytest -v -s quantization/ --ignore quantization/test_blackwell_moe.py --ignore quantization/test_rocm_moe.py --ignore quantization/test_gfx950_moe.py --shard-id=$$BUILDKITE_PARALLEL_JOB --num-shards=$$BUILDKITE_PARALLEL_JOB_COUNT
|
|
- if [ "$$BUILDKITE_PARALLEL_JOB" = "0" ]; then pytest -v -s rocm/test_moe_weight_replay.py; fi
|
|
|
|
#--------------------------------------------------------- mi355 · samplers ----------------------------------------------------------#
|
|
|
|
- label: ":amd: (MI355 DPX) Samplers Multimodal Beam Search"
|
|
timeout_in_minutes: 45
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx950nightly, amdmi355]
|
|
dind: false
|
|
agent_pool: mi355_dpx
|
|
num_gpus: 1
|
|
optional: true
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- vllm/model_executor/layers
|
|
- vllm/sampling_metadata.py
|
|
- vllm/v1/sample/
|
|
- vllm/entrypoints/generate/beam_search/
|
|
- tests/samplers
|
|
- tests/conftest.py
|
|
- vllm/_aiter_ops.py
|
|
- vllm/platforms/rocm.py
|
|
commands:
|
|
- pytest -v -s samplers/test_beam_search.py::test_beam_search_passes_multimodal_data
|
|
|
|
#------------------------------------------------------------ mi355 · v1 -------------------------------------------------------------#
|
|
|
|
- label: ":amd: (MI355) Spec Decode AL DFlash2 Nightly"
|
|
timeout_in_minutes: 60
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx950nightly, amdmi355]
|
|
dind: false
|
|
agent_pool: mi355_1
|
|
num_gpus: 1
|
|
optional: true
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- .buildkite/test_areas/spec_decode.yaml
|
|
- tests/v1/e2e/spec_decode/
|
|
- tests/evals/gsm8k/
|
|
- tests/utils.py
|
|
- vllm/v1/spec_decode/
|
|
- vllm/v1/worker/gpu/spec_decode/
|
|
- vllm/v1/worker/gpu/sample/
|
|
- vllm/v1/worker/gpu/model_runner.py
|
|
- vllm/v1/attention/
|
|
- vllm/model_executor/models/qwen3_dflash.py
|
|
- vllm/model_executor/models/qwen3_dflash2.py
|
|
- vllm/model_executor/models/qwen3_5.py
|
|
- vllm/model_executor/model_loader/
|
|
- vllm/model_executor/layers/quantization/
|
|
- vllm/model_executor/kernels/linear/
|
|
- vllm/platforms/rocm.py
|
|
commands:
|
|
- pytest -v -s v1/e2e/spec_decode/acceptance_rates/dflash/ -k "dflash2"
|
|
|
|
- label: ":amd: (MI355 DPX) Spec Decode Draft Model" # TBD
|
|
timeout_in_minutes: 60
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx950nightly, amdmi355]
|
|
dind: false
|
|
agent_pool: mi355_dpx
|
|
optional: true
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- vllm/v1/spec_decode/
|
|
- vllm/v1/worker/gpu/spec_decode/
|
|
- vllm/model_executor/model_loader/
|
|
- vllm/v1/sample/
|
|
- vllm/model_executor/layers/
|
|
- tests/v1/e2e/spec_decode/
|
|
- vllm/platforms/rocm.py
|
|
commands:
|
|
- pytest -v -s v1/e2e/spec_decode/draft_model/
|
|
|
|
- label: ":amd: (MI355 DPX) Spec Decode Eagle 1: DeepSeek + Qwen"
|
|
timeout_in_minutes: 60
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx950nightly, amdmi355]
|
|
dind: false
|
|
agent_pool: mi355_dpx
|
|
optional: true
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- vllm/v1/spec_decode/
|
|
- vllm/v1/worker/gpu/spec_decode/
|
|
- vllm/model_executor/model_loader/
|
|
- vllm/v1/sample/
|
|
- vllm/model_executor/layers/
|
|
- tests/v1/e2e/spec_decode/
|
|
- vllm/platforms/rocm.py
|
|
commands:
|
|
- pytest -v -s v1/e2e/spec_decode/eagle/ -k "deepseek_eagle or qwen3_eagle3"
|
|
|
|
- label: ":amd: (MI355 DPX) Spec Decode Eagle 2: Llama 3 + Qwen VL + Other"
|
|
timeout_in_minutes: 60
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx950nightly, amdmi355]
|
|
dind: false
|
|
agent_pool: mi355_dpx
|
|
optional: true
|
|
working_dir: /vllm-workspace/tests
|
|
source_file_dependencies:
|
|
- vllm/v1/spec_decode/
|
|
- vllm/v1/worker/gpu/spec_decode/
|
|
- vllm/model_executor/model_loader/
|
|
- vllm/v1/sample/
|
|
- vllm/model_executor/layers/
|
|
- tests/v1/e2e/spec_decode/
|
|
- vllm/platforms/rocm.py
|
|
commands:
|
|
- pytest -v -s v1/e2e/spec_decode/eagle/ -k "not deepseek_eagle and not qwen3_eagle3"
|
|
|
|
|
|
- label: ":amd: (MI355 DPX) Spec Decode N-Gram + Suffix" # TBD
|
|
timeout_in_minutes: 35
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx950nightly, amdmi355]
|
|
dind: false
|
|
agent_pool: mi355_dpx
|
|
optional: true
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- vllm/v1/spec_decode/
|
|
- vllm/v1/worker/gpu/spec_decode/
|
|
- vllm/model_executor/model_loader/
|
|
- vllm/v1/sample/
|
|
- vllm/model_executor/layers/
|
|
- tests/v1/e2e/spec_decode/
|
|
- tests/spec_decode/
|
|
- vllm/platforms/rocm.py
|
|
commands:
|
|
- pytest -v -s v1/e2e/spec_decode/ngram_suffix/
|
|
- python3 spec_decode/test_custom_proposer.py
|
|
|
|
- label: ":amd: (MI355 DPX) V1 Core"
|
|
timeout_in_minutes: 60
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx950nightly, amdmi355]
|
|
dind: false
|
|
agent_pool: mi355_dpx
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- vllm/
|
|
- tests/v1/core
|
|
- tests/v1/executor
|
|
- tests/v1/kv_offload
|
|
- tests/v1/simple_kv_offload
|
|
- tests/v1/worker
|
|
- tests/v1/streaming_input
|
|
- tests/v1/kv_connector/unit
|
|
- tests/v1/ec_connector/unit
|
|
- tests/v1/metrics
|
|
- tests/entrypoints/openai/correctness/test_lmeval.py
|
|
- vllm/config/
|
|
- vllm/distributed/
|
|
- vllm/engine/
|
|
- vllm/entrypoints/pooling/
|
|
- vllm/inputs/
|
|
- vllm/lora/
|
|
- vllm/model_executor/
|
|
- vllm/multimodal/
|
|
- vllm/outputs.py
|
|
- vllm/platforms/
|
|
- vllm/pooling_params.py
|
|
- vllm/profiler/
|
|
- vllm/sampling_params.py
|
|
- vllm/tokenizers/
|
|
- vllm/transformers_utils/
|
|
- vllm/utils/
|
|
- vllm/v1/
|
|
optional: true
|
|
commands:
|
|
# - export HSA_NO_SCRATCH_RECLAIM=1
|
|
- export VLLM_WORKER_MULTIPROC_METHOD=spawn
|
|
- pytest -v -s -m 'not cpu_test' v1/core
|
|
|
|
- label: ":amd: (MI355 DPX) V1 Executor + Worker"
|
|
timeout_in_minutes: 60
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx950nightly, amdmi355]
|
|
dind: false
|
|
agent_pool: mi355_dpx
|
|
working_dir: /vllm-workspace/tests
|
|
source_file_dependencies:
|
|
- vllm/
|
|
- tests/v1/core
|
|
- tests/v1/executor
|
|
- tests/v1/kv_offload
|
|
- tests/v1/simple_kv_offload
|
|
- tests/v1/worker
|
|
- tests/v1/streaming_input
|
|
- tests/v1/kv_connector/unit
|
|
- tests/v1/ec_connector/unit
|
|
- tests/v1/metrics
|
|
- tests/entrypoints/openai/correctness/test_lmeval.py
|
|
- vllm/config/
|
|
- vllm/distributed/
|
|
- vllm/engine/
|
|
- vllm/entrypoints/pooling/
|
|
- vllm/inputs/
|
|
- vllm/lora/
|
|
- vllm/model_executor/
|
|
- vllm/multimodal/
|
|
- vllm/outputs.py
|
|
- vllm/platforms/
|
|
- vllm/pooling_params.py
|
|
- vllm/profiler/
|
|
- vllm/sampling_params.py
|
|
- vllm/tokenizers/
|
|
- vllm/transformers_utils/
|
|
- vllm/utils/
|
|
- vllm/v1/
|
|
commands:
|
|
- export VLLM_WORKER_MULTIPROC_METHOD=spawn
|
|
- pytest -v -s v1/executor
|
|
- pytest -v -s v1/worker
|
|
- pytest -v -s v1/streaming_input
|
|
optional: true
|
|
|
|
- label: ":amd: (MI355 DPX) V1 KV Offload"
|
|
timeout_in_minutes: 60
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx950nightly, amdmi355]
|
|
dind: false
|
|
agent_pool: mi355_dpx
|
|
working_dir: /vllm-workspace/tests
|
|
source_file_dependencies:
|
|
- vllm/
|
|
- tests/v1/core
|
|
- tests/v1/executor
|
|
- tests/v1/kv_offload
|
|
- tests/v1/simple_kv_offload
|
|
- tests/v1/worker
|
|
- tests/v1/streaming_input
|
|
- tests/v1/kv_connector/unit
|
|
- tests/v1/ec_connector/unit
|
|
- tests/v1/metrics
|
|
- tests/entrypoints/openai/correctness/test_lmeval.py
|
|
- vllm/config/
|
|
- vllm/distributed/
|
|
- vllm/engine/
|
|
- vllm/entrypoints/pooling/
|
|
- vllm/inputs/
|
|
- vllm/lora/
|
|
- vllm/model_executor/
|
|
- vllm/multimodal/
|
|
- vllm/outputs.py
|
|
- vllm/platforms/
|
|
- vllm/pooling_params.py
|
|
- vllm/profiler/
|
|
- vllm/sampling_params.py
|
|
- vllm/tokenizers/
|
|
- vllm/transformers_utils/
|
|
- vllm/utils/
|
|
- vllm/v1/
|
|
commands:
|
|
- export VLLM_WORKER_MULTIPROC_METHOD=spawn
|
|
- bash /vllm-workspace/.buildkite/scripts/install-kv-offload.sh
|
|
- pytest -v -s v1/kv_offload
|
|
- pytest -v -s v1/simple_kv_offload
|
|
optional: true
|
|
|
|
- label: ":amd: (MI355 DPX) V1 KV Connectors Shard %N"
|
|
timeout_in_minutes: 60
|
|
parallelism: 4
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx950nightly, amdmi355]
|
|
dind: false
|
|
agent_pool: mi355_dpx
|
|
working_dir: /vllm-workspace/tests
|
|
source_file_dependencies:
|
|
- vllm/
|
|
- tests/v1/core
|
|
- tests/v1/executor
|
|
- tests/v1/kv_offload
|
|
- tests/v1/simple_kv_offload
|
|
- tests/v1/worker
|
|
- tests/v1/streaming_input
|
|
- tests/v1/kv_connector/unit
|
|
- tests/v1/ec_connector/unit
|
|
- tests/v1/metrics
|
|
- tests/entrypoints/openai/correctness/test_lmeval.py
|
|
- vllm/config/
|
|
- vllm/distributed/
|
|
- vllm/engine/
|
|
- vllm/entrypoints/pooling/
|
|
- vllm/inputs/
|
|
- vllm/lora/
|
|
- vllm/model_executor/
|
|
- vllm/multimodal/
|
|
- vllm/outputs.py
|
|
- vllm/platforms/
|
|
- vllm/pooling_params.py
|
|
- vllm/profiler/
|
|
- vllm/sampling_params.py
|
|
- vllm/tokenizers/
|
|
- vllm/transformers_utils/
|
|
- vllm/utils/
|
|
- vllm/v1/
|
|
commands:
|
|
- bash /vllm-workspace/.buildkite/scripts/install-kv-connectors.sh
|
|
- export VLLM_WORKER_MULTIPROC_METHOD=spawn
|
|
- pytest -v -s -m 'not cpu_test' v1/kv_connector/unit --num-shards=$$BUILDKITE_PARALLEL_JOB_COUNT --shard-id=$$BUILDKITE_PARALLEL_JOB
|
|
- pytest -v -s -m 'not cpu_test' v1/ec_connector/unit --num-shards=$$BUILDKITE_PARALLEL_JOB_COUNT --shard-id=$$BUILDKITE_PARALLEL_JOB
|
|
optional: true
|
|
|
|
- label: ":amd: (MI355 DPX) V1 Metrics + LM Eval"
|
|
timeout_in_minutes: 65
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx950nightly, amdmi355]
|
|
dind: false
|
|
agent_pool: mi355_dpx
|
|
working_dir: /vllm-workspace/tests
|
|
source_file_dependencies:
|
|
- vllm/
|
|
- tests/v1/core
|
|
- tests/v1/executor
|
|
- tests/v1/kv_offload
|
|
- tests/v1/simple_kv_offload
|
|
- tests/v1/worker
|
|
- tests/v1/streaming_input
|
|
- tests/v1/kv_connector/unit
|
|
- tests/v1/ec_connector/unit
|
|
- tests/v1/metrics
|
|
- tests/entrypoints/openai/correctness/test_lmeval.py
|
|
- vllm/config/
|
|
- vllm/distributed/
|
|
- vllm/engine/
|
|
- vllm/entrypoints/pooling/
|
|
- vllm/inputs/
|
|
- vllm/lora/
|
|
- vllm/model_executor/
|
|
- vllm/multimodal/
|
|
- vllm/outputs.py
|
|
- vllm/platforms/
|
|
- vllm/pooling_params.py
|
|
- vllm/profiler/
|
|
- vllm/sampling_params.py
|
|
- vllm/tokenizers/
|
|
- vllm/transformers_utils/
|
|
- vllm/utils/
|
|
- vllm/v1/
|
|
commands:
|
|
- export GIT_CONFIG_COUNT=1
|
|
- export GIT_CONFIG_KEY_0=http.version
|
|
- export GIT_CONFIG_VALUE_0=HTTP/1.1
|
|
- export GIT_TERMINAL_PROMPT=0
|
|
- export VLLM_WORKER_MULTIPROC_METHOD=spawn
|
|
- pytest -v -s -m 'not cpu_test' v1/metrics
|
|
- pip install -U git+https://github.com/vllm-project/lm-evaluation-harness.git@streaming-api
|
|
- pytest -v -s entrypoints/openai/correctness/test_lmeval.py::test_lm_eval_accuracy_v1_engine
|
|
optional: true
|
|
|
|
|
|
- label: ":amd: (MI355 DPX) V1 Sample"
|
|
timeout_in_minutes: 70
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx950nightly, amdmi355]
|
|
dind: false
|
|
agent_pool: mi355_dpx
|
|
optional: true
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- vllm/
|
|
- tests/v1/sample
|
|
- tests/v1/logits_processors
|
|
- tests/v1/test_oracle.py
|
|
- tests/v1/test_request.py
|
|
- tests/v1/test_outputs.py
|
|
- vllm/config/
|
|
- vllm/distributed/
|
|
- vllm/engine/
|
|
- vllm/inputs/
|
|
- vllm/logger.py
|
|
- vllm/model_executor/
|
|
- vllm/platforms/
|
|
- vllm/sampling_params.py
|
|
- vllm/transformers_utils/
|
|
- vllm/utils/
|
|
- vllm/v1/
|
|
commands:
|
|
- export VLLM_WORKER_MULTIPROC_METHOD=spawn
|
|
- pytest -v -s v1/sample
|
|
|
|
- label: ":amd: (MI355 DPX) V1 Logits + Oracle"
|
|
timeout_in_minutes: 70
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx950nightly, amdmi355]
|
|
dind: false
|
|
agent_pool: mi355_dpx
|
|
optional: true
|
|
working_dir: /vllm-workspace/tests
|
|
source_file_dependencies:
|
|
- vllm/
|
|
- tests/v1/sample
|
|
- tests/v1/logits_processors
|
|
- tests/v1/test_oracle.py
|
|
- tests/v1/test_request.py
|
|
- tests/v1/test_outputs.py
|
|
- vllm/config/
|
|
- vllm/distributed/
|
|
- vllm/engine/
|
|
- vllm/inputs/
|
|
- vllm/logger.py
|
|
- vllm/model_executor/
|
|
- vllm/platforms/
|
|
- vllm/sampling_params.py
|
|
- vllm/transformers_utils/
|
|
- vllm/utils/
|
|
- vllm/v1/
|
|
commands:
|
|
- export VLLM_WORKER_MULTIPROC_METHOD=spawn
|
|
- pytest -v -s v1/logits_processors
|
|
- pytest -v -s v1/test_oracle.py
|
|
- pytest -v -s v1/test_request.py
|
|
- pytest -v -s v1/test_outputs.py
|
|
|
|
|
|
- label: ":amd: (MI355 DPX) E2E Core Large Memory"
|
|
timeout_in_minutes: 65
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx950nightly, amdmi355]
|
|
dind: false
|
|
agent_pool: mi355_dpx
|
|
num_gpus: 1
|
|
optional: true
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- vllm/v1/
|
|
- tests/v1/e2e/general/
|
|
- vllm/platforms/rocm.py
|
|
commands:
|
|
- pytest -v -s v1/e2e/general/test_kv_sharing_fast_prefill.py
|
|
- pytest -v -s v1/e2e/general/test_mamba_prefix_cache.py::test_mamba_prefix_cache_mrv2 v1/e2e/general/test_mamba_prefix_cache.py::test_mamba_prefix_cache_mrv2_async
|
|
|
|
- label: ":amd: (MI355 DPX) Extract Hidden States Integration Single GPU"
|
|
timeout_in_minutes: 45
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx950nightly, amdmi355]
|
|
dind: false
|
|
agent_pool: mi355_dpx
|
|
num_gpus: 1
|
|
optional: true
|
|
working_dir: "/vllm-workspace"
|
|
source_file_dependencies:
|
|
- vllm/v1/spec_decode/extract_hidden_states.py
|
|
- vllm/model_executor/models/extract_hidden_states.py
|
|
- vllm/transformers_utils/configs/extract_hidden_states.py
|
|
- vllm/distributed/kv_transfer/kv_connector/v1/example_hidden_states_connector.py
|
|
- tests/v1/kv_connector/extract_hidden_states_integration
|
|
- vllm/platforms/rocm.py
|
|
commands:
|
|
- export VLLM_WORKER_MULTIPROC_METHOD=spawn
|
|
- pytest -v -s -m 'not distributed' tests/v1/kv_connector/extract_hidden_states_integration
|
|
|
|
- label: ":amd: (MI355 DPX) V1 Attention Shard %N" # TBD
|
|
timeout_in_minutes: 125
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx950nightly, amdmi355]
|
|
dind: false
|
|
agent_pool: mi355_dpx
|
|
parallelism: 2
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- vllm/config/attention.py
|
|
- vllm/model_executor/layers/attention
|
|
- vllm/v1/attention
|
|
- tests/v1/attention
|
|
- vllm/_aiter_ops.py
|
|
- vllm/envs.py
|
|
- vllm/platforms/rocm.py
|
|
optional: true
|
|
commands:
|
|
- pytest -v -s v1/attention --ignore=v1/attention/test_mla_backends.py --shard-id=$$BUILDKITE_PARALLEL_JOB --num-shards=$$BUILDKITE_PARALLEL_JOB_COUNT
|
|
- VLLM_ROCM_USE_AITER=1 pytest -v -s v1/attention/test_mla_backends.py --shard-id=$$BUILDKITE_PARALLEL_JOB --num-shards=$$BUILDKITE_PARALLEL_JOB_COUNT
|
|
|
|
- label: ":amd: (MI355 DPX) CUDAGraph"
|
|
timeout_in_minutes: 55
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx950nightly, amdmi355]
|
|
dind: false
|
|
agent_pool: mi355_dpx
|
|
optional: true
|
|
working_dir: /vllm-workspace/tests
|
|
source_file_dependencies:
|
|
- vllm/
|
|
- tests/v1/core
|
|
- tests/v1/executor
|
|
- tests/v1/kv_offload
|
|
- tests/v1/worker
|
|
- tests/v1/cudagraph
|
|
- tests/v1/kv_connector/unit
|
|
- tests/v1/metrics
|
|
- tests/entrypoints/openai/correctness/test_lmeval.py
|
|
- vllm/v1/cudagraph_dispatcher.py
|
|
- vllm/config/compilation.py
|
|
- vllm/compilation
|
|
- vllm/platforms/rocm.py
|
|
- vllm/v1/worker/encoder_cudagraph.py
|
|
- vllm/v1/worker/encoder_cudagraph_defs.py
|
|
commands:
|
|
- pytest -v -s v1/cudagraph/test_cudagraph_dispatch.py
|
|
- pytest -v -s v1/cudagraph/test_cudagraph_mode.py
|
|
- pytest -v -s v1/cudagraph/test_breakable_cudagraph.py
|
|
- pytest -v -s v1/cudagraph/test_encoder_cudagraph.py
|
|
|
|
|
|
- label: ":amd: (MI355 DPX) V1 Spec Decode" # TBD
|
|
timeout_in_minutes: 50
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx950nightly, amdmi355]
|
|
dind: false
|
|
agent_pool: mi355_dpx
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- vllm/
|
|
- tests/v1/spec_decode
|
|
- vllm/config/
|
|
- vllm/distributed/
|
|
- vllm/inputs/
|
|
- vllm/model_executor/
|
|
- vllm/models/qwen4_exp/
|
|
- vllm/platforms/
|
|
- vllm/sampling_params.py
|
|
- vllm/transformers_utils/
|
|
- vllm/utils/
|
|
- vllm/v1/
|
|
optional: true
|
|
commands:
|
|
- export VLLM_WORKER_MULTIPROC_METHOD=spawn
|
|
- pytest -v -s -m 'not slow_test' v1/spec_decode
|
|
|
|
- label: ":amd: (MI355 DPX) Spec Decode AL MTP + Other Acceptance Nightly"
|
|
timeout_in_minutes: 55
|
|
parallelism: 3
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx950nightly, amdmi355]
|
|
dind: false
|
|
agent_pool: mi355_dpx
|
|
optional: true
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- vllm/v1/spec_decode/
|
|
- vllm/v1/sample/
|
|
- vllm/v1/worker/gpu/
|
|
- vllm/v1/worker/gpu_model_runner.py
|
|
- vllm/v1/worker/gpu_worker.py
|
|
- vllm/model_executor/model_loader/
|
|
- vllm/model_executor/models/gemma4.py
|
|
- vllm/model_executor/models/gemma4_mm.py
|
|
- vllm/model_executor/models/gemma4_mtp.py
|
|
- vllm/model_executor/models/llama.py
|
|
- vllm/model_executor/models/llama_eagle3.py
|
|
- vllm/model_executor/models/medusa.py
|
|
- vllm/model_executor/models/registry.py
|
|
- tests/evals/gsm8k/
|
|
- tests/utils.py
|
|
- tests/v1/e2e/spec_decode/
|
|
- vllm/platforms/rocm.py
|
|
- vllm/v1/worker/gpu/spec_decode/
|
|
- tests/v1/e2e/spec_decode/conftest.py
|
|
commands:
|
|
- if [ "$$BUILDKITE_PARALLEL_JOB" = "0" ]; then pytest -v -s v1/e2e/spec_decode/acceptance_rates/mtp_other/ --deselect "tests/v1/e2e/spec_decode/acceptance_rates/mtp_other/test_mtp.py::test_gemma4_mtp_acceptance_lengths[False]" --deselect "tests/v1/e2e/spec_decode/acceptance_rates/mtp_other/test_mtp.py::test_gemma4_mtp_acceptance_lengths[True]"; fi
|
|
- if [ "$$BUILDKITE_PARALLEL_JOB" = "1" ]; then pytest -v -s "v1/e2e/spec_decode/acceptance_rates/mtp_other/test_mtp.py::test_gemma4_mtp_acceptance_lengths[False]"; fi
|
|
- if [ "$$BUILDKITE_PARALLEL_JOB" = "2" ]; then pytest -v -s "v1/e2e/spec_decode/acceptance_rates/mtp_other/test_mtp.py::test_gemma4_mtp_acceptance_lengths[True]"; fi
|
|
|
|
#----------------------------------------------------------- mi355 · misc ------------------------------------------------------------#
|
|
|
|
- label: ":amd: (MI355 DPX) Regression" # TBD
|
|
timeout_in_minutes: 25
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx950nightly, amdmi355]
|
|
dind: false
|
|
agent_pool: mi355_dpx
|
|
optional: true
|
|
working_dir: "/vllm-workspace/tests"
|
|
source_file_dependencies:
|
|
- vllm/
|
|
- tests/test_regression
|
|
- vllm/config/
|
|
- vllm/distributed/
|
|
- vllm/engine/
|
|
- vllm/inputs/
|
|
- vllm/model_executor/
|
|
- vllm/multimodal/
|
|
- vllm/platforms/
|
|
- vllm/sampling_params.py
|
|
- vllm/transformers_utils/
|
|
- vllm/utils/
|
|
- vllm/v1/
|
|
commands:
|
|
- pip install 'modelscope<1.38'
|
|
- pytest -v -s test_regression.py
|
|
|
|
- label: ':amd: (MI355) DeepSeek V4 MoE Integration'
|
|
timeout_in_minutes: 20
|
|
source_file_dependencies:
|
|
- tests/models/test_deepseek_v4_vl_rocm.py
|
|
- vllm/models/deepseek_v4/amd/
|
|
- vllm/models/deepseek_v4/common/
|
|
- vllm/model_executor/layers/fused_moe/
|
|
- vllm/model_executor/layers/activation.py
|
|
- vllm/model_executor/layers/linear.py
|
|
- vllm/config/
|
|
- vllm/forward_context.py
|
|
- vllm/v1/worker/workspace.py
|
|
- vllm/_aiter_ops.py
|
|
- vllm/platforms/rocm.py
|
|
- vllm/envs.py
|
|
commands:
|
|
- VLLM_ROCM_USE_AITER_MOE=0 pytest -v -s models/test_deepseek_v4_vl_rocm.py
|
|
dind: false
|
|
working_dir: /vllm-workspace/tests
|
|
agent_pool: mi355_1
|
|
num_gpus: 1
|
|
mirror_hardwares:
|
|
- amdexperimental
|
|
- amdproduction
|
|
- amdgfx950nightly
|
|
- amdmi355
|
|
optional: true
|
|
|
|
- label: ':amd: (MI355) Quantized MoE'
|
|
timeout_in_minutes: 45
|
|
working_dir: /vllm-workspace/
|
|
source_file_dependencies:
|
|
- tests/quantization/test_gfx950_moe.py
|
|
- tests/quantization/test_rocm_moe.py
|
|
- tests/utils.py
|
|
- vllm/model_executor/models/deepseek_v2.py
|
|
- vllm/model_executor/models/qwen3_moe.py
|
|
- vllm/model_executor/models/gpt_oss.py
|
|
- vllm/model_executor/model_loader/
|
|
- vllm/model_executor/layers/fused_moe/
|
|
- vllm/model_executor/layers/quantization/
|
|
- vllm/model_executor/kernels/linear/
|
|
- vllm/v1/attention/
|
|
- vllm/v1/worker/gpu/
|
|
- vllm/_aiter_ops.py
|
|
- vllm/platforms/rocm.py
|
|
- vllm/envs.py
|
|
commands:
|
|
- pytest -v -s tests/quantization/test_rocm_moe.py
|
|
- pytest -v -s tests/quantization/test_gfx950_moe.py
|
|
dind: false
|
|
agent_pool: mi355_1
|
|
num_gpus: 1
|
|
mirror_hardwares:
|
|
- amdexperimental
|
|
- amdproduction
|
|
- amdgfx950nightly
|
|
- amdmi355
|
|
optional: true
|
|
|
|
- label: ':amd: (MI355) Fusion E2E TP2 AsyncTP BF16 Config Sweep'
|
|
timeout_in_minutes: 30
|
|
working_dir: /vllm-workspace/
|
|
source_file_dependencies:
|
|
- vllm/compilation/
|
|
- vllm/model_executor/layers/linear.py
|
|
- vllm/model_executor/layers/utils.py
|
|
- vllm/distributed/
|
|
- vllm/v1/attention/backends/triton_attn.py
|
|
- tests/compile/fusions_e2e/
|
|
- tests/compile/passes/distributed/test_async_tp.py
|
|
- tests/compile/rocm/test_async_tp.py
|
|
- tests/compile/correctness_e2e/test_async_tp.py
|
|
- vllm/model_executor/kernels/linear/
|
|
- vllm/model_executor/models/qwen3.py
|
|
commands:
|
|
- rocm-smi
|
|
- pytest -v -s tests/compile/rocm/test_async_tp.py
|
|
- pytest -v -s tests/compile/fusions_e2e/test_tp2_rocm_async_tp.py
|
|
- pytest -v -s tests/compile/correctness_e2e/test_async_tp.py::test_rocm_async_tp_bf16_output_correctness
|
|
dind: false
|
|
agent_pool: mi355_2
|
|
num_gpus: 2
|
|
mirror_hardwares:
|
|
- amdexperimental
|
|
- amdproduction
|
|
- amdgfx950nightly
|
|
- amdmi355
|
|
optional: true
|
|
|
|
- label: ":amd: (MI355 DPX) MRV2 Sampler JIT Warmup"
|
|
timeout_in_minutes: 10
|
|
mirror_hardwares: [amdexperimental, amdproduction, amdgfx950nightly, amdmi355]
|
|
dind: false
|
|
agent_pool: mi355_dpx
|
|
num_gpus: 1
|
|
optional: true
|
|
working_dir: "/vllm-workspace/"
|
|
source_file_dependencies:
|
|
- tests/jit_monitor/
|
|
- tests/models/utils.py
|
|
- tests/models/registry.py
|
|
- tests/utils.py
|
|
- vllm/v1/worker/gpu/
|
|
- vllm/v1/worker/gpu_worker.py
|
|
- vllm/model_executor/warmup/
|
|
- vllm/v1/sample/ops/topk_topp_sampler.py
|
|
- vllm/utils/jit_monitor.py
|
|
- vllm/config/observability.py
|
|
- vllm/platforms/rocm.py
|
|
- vllm/envs.py
|
|
commands:
|
|
- pytest -v -s tests/jit_monitor/test_no_runtime_jit_rocm.py --timeout=300
|