diff --git a/.buildkite/ci_config.yaml b/.buildkite/ci_config.yaml index d694d2910b6f..3fef1d30342c 100644 --- a/.buildkite/ci_config.yaml +++ b/.buildkite/ci_config.yaml @@ -1,4 +1,7 @@ name: vllm_ci +# Every eligible step shards from main's timings (ci-infra runtime_shard.py); +# `automatic_shard: false` on a step opts it out. +automatic_shard_default: true job_dirs: - ".buildkite/image_build" - ".buildkite/test_areas" diff --git a/.buildkite/test_areas/attention.yaml b/.buildkite/test_areas/attention.yaml index d1ca74b9f40b..b7f6dfa3ca61 100644 --- a/.buildkite/test_areas/attention.yaml +++ b/.buildkite/test_areas/attention.yaml @@ -4,7 +4,6 @@ depends_on: steps: - label: ":nvidia: (H100) V1 Attention" key: v1-attention-h100-mi300 - automatic_shard: true timeout_in_minutes: 85 device: h100 source_file_dependencies: @@ -22,7 +21,6 @@ steps: amd: label: ":amd: (MI300) V1 Attention" dind: false - automatic_shard: true device: mi300_1 timeout_in_minutes: 125 depends_on: @@ -53,7 +51,6 @@ steps: - label: ":nvidia: (B200) V1 Attention" key: v1-attention-b200 - automatic_shard: true timeout_in_minutes: 80 device: b200-k8s source_file_dependencies: @@ -70,7 +67,6 @@ steps: amd: label: ":amd: (MI355 DPX) V1 Attention" dind: false - automatic_shard: true device: mi355_dpx timeout_in_minutes: 125 depends_on: @@ -101,7 +97,6 @@ steps: - label: ":nvidia: (DGX Spark) [Nightly] FlashInfer NVFP4 XQA" key: flashinfer-nvfp4-xqa-dgx-spark - automatic_shard: true timeout_in_minutes: 30 device: dgx-spark optional: true diff --git a/.buildkite/test_areas/basic_correctness.yaml b/.buildkite/test_areas/basic_correctness.yaml index 51908178f7d9..c8b26d40b850 100644 --- a/.buildkite/test_areas/basic_correctness.yaml +++ b/.buildkite/test_areas/basic_correctness.yaml @@ -4,7 +4,6 @@ depends_on: steps: - label: ":nvidia: (H200 MIG 18GB) Basic Correctness Models" key: basic-correctness - automatic_shard: true timeout_in_minutes: 68 device: h200_18gb source_file_dependencies: &basic-correctness-deps @@ -17,7 +16,6 @@ steps: mirror: amd: label: ":amd: (MI355 DPX) Basic Correctness Models" - automatic_shard: true dind: false device: mi355_dpx timeout_in_minutes: 60 @@ -26,7 +24,6 @@ steps: - label: ":nvidia: (H200 MIG 18GB) Basic Correctness CuMem" key: basic-correctness-cumem - automatic_shard: true timeout_in_minutes: 68 device: h200_18gb source_file_dependencies: *basic-correctness-deps @@ -35,7 +32,6 @@ steps: - pytest -v -s basic_correctness/memory/cumem mirror: amd: - automatic_shard: true label: ":amd: (MI355 DPX) Basic Correctness CuMem" dind: false device: mi355_dpx @@ -54,7 +50,7 @@ steps: - label: ":nvidia: (H200 MIG 18GB) Basic Correctness Sleep Mode" key: basic-correctness-sleep-mode - automatic_shard: true + automatic_shard: false # staging: a step that opts out, and the mirror that follows it timeout_in_minutes: 68 device: h200_18gb source_file_dependencies: *basic-correctness-deps @@ -63,7 +59,6 @@ steps: - pytest -v -s basic_correctness/memory/sleep_mode mirror: amd: - automatic_shard: true label: ":amd: (MI300 DPX) Basic Correctness Sleep Mode" dind: false device: mi300_dpx @@ -73,7 +68,6 @@ steps: - label: ":nvidia: (H200 MIG 18GB) Basic Correctness CPU Offload" key: basic-correctness-cpu-offload - automatic_shard: true timeout_in_minutes: 68 device: h200_18gb source_file_dependencies: *basic-correctness-deps @@ -82,7 +76,6 @@ steps: - pytest -v -s basic_correctness/cpu_offload mirror: amd: - automatic_shard: true label: ":amd: (MI355 DPX) Basic Correctness CPU Offload" dind: false device: mi355_dpx @@ -92,7 +85,6 @@ steps: - label: ":nvidia: (H200 MIG 18GB) Basic Correctness Prefetch Offload" key: basic-correctness-prefetch-offload - automatic_shard: true timeout_in_minutes: 68 device: h200_18gb source_file_dependencies: *basic-correctness-deps @@ -101,7 +93,6 @@ steps: - pytest -v -s basic_correctness/prefetch_offload mirror: amd: - automatic_shard: true label: ":amd: (MI355 DPX) Basic Correctness Prefetch Offload" dind: false device: mi355_dpx diff --git a/.buildkite/test_areas/benchmarks.yaml b/.buildkite/test_areas/benchmarks.yaml index 1eca056cbb9c..4df7a136f059 100644 --- a/.buildkite/test_areas/benchmarks.yaml +++ b/.buildkite/test_areas/benchmarks.yaml @@ -4,7 +4,6 @@ depends_on: steps: - label: ":nvidia: (H200 MIG 18GB) Benchmarks CLI" key: benchmarks-cli-test - automatic_shard: true timeout_in_minutes: 55 device: h200_18gb source_file_dependencies: @@ -16,7 +15,6 @@ steps: - pytest -v -s benchmarks/ mirror: amd: - automatic_shard: true label: ":amd: (MI355 DPX) Benchmarks CLI" dind: false device: mi355_dpx diff --git a/.buildkite/test_areas/compile.yaml b/.buildkite/test_areas/compile.yaml index 878c54a149c2..af1edb9ddd85 100644 --- a/.buildkite/test_areas/compile.yaml +++ b/.buildkite/test_areas/compile.yaml @@ -4,7 +4,6 @@ depends_on: steps: - label: ":nvidia: (H100) Distributed Compile" key: distributed-compile-unit-tests-2xh100 - automatic_shard: true timeout_in_minutes: 45 working_dir: "/vllm-workspace/" device: h100 @@ -18,7 +17,6 @@ steps: - pytest -s -v tests/compile/passes/distributed mirror: amd: - automatic_shard: true label: ':amd: (MI300) Distributed Compile' dind: false device: mi300_2 @@ -38,7 +36,6 @@ steps: - label: ":nvidia: (B200) Fusion and Compile" key: fusion-and-compile-unit-tests-2xb200 - automatic_shard: true timeout_in_minutes: 30 working_dir: "/vllm-workspace/" device: b200-k8s @@ -68,7 +65,6 @@ steps: - pytest -v -s tests/compile/passes/distributed/test_fusion_all_reduce.py mirror: amd: - automatic_shard: true label: ":amd: (MI355) Fusion and Compile FP8" dind: false device: mi355_2 @@ -99,7 +95,6 @@ steps: - label: ":nvidia: (H100) Fusion E2E Quick" key: fusion-e2e-quick-h100 - automatic_shard: true timeout_in_minutes: 25 working_dir: "/vllm-workspace/" device: h100 @@ -118,7 +113,6 @@ steps: - pytest -v -s tests/compile/fusions_e2e/test_tp1_quant.py -k "inductor_partition and not +rms_norm and +quant_fp8 and (qwen3 or deepseek)" mirror: amd: - automatic_shard: true label: ":amd: (MI355 DPX) Fusion E2E Quick" dind: false device: mi355_dpx @@ -141,7 +135,6 @@ steps: - label: ":nvidia: (H100) [Nightly] Fusion E2E Config Sweep" key: fusion-e2e-config-sweep-h100 - automatic_shard: true timeout_in_minutes: 25 working_dir: "/vllm-workspace/" device: h100 @@ -164,7 +157,6 @@ steps: amd: # Run on PRs even when the H100 step is optional (nightly only). optional: false - automatic_shard: true label: ":amd: (MI355 DPX) Fusion E2E Config Sweep" dind: false device: mi355_dpx @@ -188,7 +180,6 @@ steps: - label: ":nvidia: (B200) [Nightly] Fusion E2E Config Sweep" key: fusion-e2e-config-sweep-b200 - automatic_shard: true timeout_in_minutes: 30 working_dir: "/vllm-workspace/" device: b200-k8s @@ -202,7 +193,6 @@ steps: - pytest -v -s tests/compile/fusions_e2e/test_tp1_quant.py -k "inductor_partition and (FLASHINFER and not +rms_norm and (not +quant_fp8 or +quant_fp8 and (qwen3 or deepseek)) or llama-3)" mirror: amd: - automatic_shard: true label: ":amd: (MI355 DPX) Fusion E2E FP8 Config Sweep" dind: false device: mi355_dpx @@ -225,7 +215,6 @@ steps: - label: ":nvidia: (H100) Fusion E2E TP2 Quick" key: fusion-e2e-tp2-quick-h100 - automatic_shard: true timeout_in_minutes: 35 working_dir: "/vllm-workspace/" device: h100 @@ -242,7 +231,6 @@ steps: - pytest -v -s tests/compile/fusions_e2e/test_tp2_ar_rms.py -k "inductor_partition and not +rms_norm and (not +quant_fp8 or +quant_fp8 and (qwen3 or deepseek))" mirror: amd: - automatic_shard: true label: ":amd: (MI355) Fusion E2E TP2 Quick" dind: false device: mi355_2 @@ -290,7 +278,6 @@ steps: - label: ":nvidia: (H100) [Nightly] Fusion E2E TP2 AR-RMS Config Sweep" key: fusion-e2e-tp2-ar-rms-config-sweep-h100 - automatic_shard: true timeout_in_minutes: 30 working_dir: "/vllm-workspace/" device: h100 @@ -313,7 +300,6 @@ steps: amd: # Run on PRs even when the H100 step is optional (nightly only). optional: false - automatic_shard: true label: ":amd: (MI355) Fusion E2E TP2 AR-RMS FP8 Config Sweep" dind: false device: mi355_2 @@ -340,7 +326,6 @@ steps: - label: ":nvidia: (B200) Fusion E2E TP2" key: fusion-e2e-tp2-b200 - automatic_shard: true timeout_in_minutes: 45 working_dir: "/vllm-workspace/" device: b200-k8s @@ -359,7 +344,6 @@ steps: - pytest -v -s tests/compile/fusions_e2e/test_tp2_ar_rms.py -k "(FLASHINFER and inductor_partition and not +rms_norm and (not +quant_fp8 or +quant_fp8 and (qwen3 or deepseek))) or Llama-3.1-8B-Instruct-FP4" mirror: amd: - automatic_shard: true label: ":amd: (MI355) Fusion E2E TP2 FP8 BF16 rewrites" dind: false device: mi355_2 @@ -388,7 +372,6 @@ steps: - label: ':amd: (MI355) [Nightly] Fusion E2E TP2 AR-RMS (Inductor Partition)' key: fusion-e2e-tp2-ar-rms-inductor-partition-amd optional: true - automatic_shard: true dind: false device: mi355_2 num_devices: 2 @@ -411,7 +394,6 @@ steps: - label: ':amd: (MI355) [Nightly] Fusion E2E TP2 AR-RMS (Dynamo Partition)' key: fusion-e2e-tp2-ar-rms-dynamo-partition-amd optional: true - automatic_shard: true dind: false device: mi355_2 num_devices: 2 diff --git a/.buildkite/test_areas/cuda.yaml b/.buildkite/test_areas/cuda.yaml index 3135b26cc03d..a2f945465d6a 100644 --- a/.buildkite/test_areas/cuda.yaml +++ b/.buildkite/test_areas/cuda.yaml @@ -4,7 +4,6 @@ depends_on: steps: - label: ":nvidia: (H200 MIG 18GB) CUDA Platform" key: platform-tests - automatic_shard: true timeout_in_minutes: 20 device: h200_18gb source_file_dependencies: @@ -20,7 +19,6 @@ steps: - pytest -v -s cuda/test_cuda_compatibility_path.py mirror: amd: - automatic_shard: true label: ":amd: (MI355 DPX) CUDA Platform" dind: false device: mi355_dpx @@ -35,7 +33,6 @@ steps: - label: ":nvidia: (H200 MIG 18GB) CUDAGraph" device: h200_18gb key: cudagraph - automatic_shard: true timeout_in_minutes: 45 source_file_dependencies: - tests/v1/cudagraph @@ -56,7 +53,6 @@ steps: - pytest -v -s v1/cudagraph/test_encoder_only_full_cudagraph.py mirror: amd: - automatic_shard: true label: ":amd: (MI250) CUDAGraph" dind: false device: mi250_1 diff --git a/.buildkite/test_areas/disaggregated.yaml b/.buildkite/test_areas/disaggregated.yaml index 84e06839004d..b22f77cbc2fd 100644 --- a/.buildkite/test_areas/disaggregated.yaml +++ b/.buildkite/test_areas/disaggregated.yaml @@ -4,7 +4,6 @@ depends_on: steps: - label: ":nvidia: (L4) [Nightly] Distributed NixlConnector PD accuracy" key: distributed-nixlconnector-pd-accuracy-4-gpus - automatic_shard: true # Nightly only: 4-GPU L4 pods wait behind smaller jobs on per-PR runs. optional: true timeout_in_minutes: 55 @@ -23,7 +22,6 @@ steps: # Run on PRs even when the L4 step is optional (nightly only). optional: false label: ":amd: (MI355) Distributed NixlConnector PD accuracy" - automatic_shard: true dind: false device: mi355_4 timeout_in_minutes: 60 @@ -39,7 +37,6 @@ steps: - label: ":nvidia: (L4) [Nightly] Distributed FlashInfer NixlConnector PD accuracy" key: distributed-flashinfer-nixlconnector-pd-accuracy-4-gpus - automatic_shard: true # Nightly only: 4-GPU L4 pods wait behind smaller jobs on per-PR runs. optional: true timeout_in_minutes: 55 @@ -58,7 +55,6 @@ steps: # Run on PRs even when the L4 step is optional (nightly only). optional: false label: ":amd: (MI355) Distributed AITER NixlConnector PD accuracy" - automatic_shard: true dind: false device: mi355_4 num_devices: 4 @@ -81,7 +77,6 @@ steps: - label: ":nvidia: (L4) [Nightly] Push NixlConnector PP prefill PD accuracy" key: push-nixlconnector-pp-prefill-pd-accuracy-4-gpus - automatic_shard: true # Nightly only: 4-GPU L4 pods wait behind smaller jobs on per-PR runs. optional: true timeout_in_minutes: 30 @@ -101,7 +96,6 @@ steps: # Run on PRs even when the L4 step is optional (nightly only). optional: false label: ":amd: (MI300) Push NixlConnector PP prefill PD accuracy" - automatic_shard: true dind: false device: mi300_4 timeout_in_minutes: 55 @@ -118,7 +112,6 @@ steps: - label: ":nvidia: (L4) [Nightly] DP EP Distributed NixlConnector PD accuracy" key: dp-ep-distributed-nixlconnector-pd-accuracy-tests-4-gpus - automatic_shard: true # Nightly only: 4-GPU L4 pods wait behind smaller jobs on per-PR runs. optional: true timeout_in_minutes: 30 @@ -137,7 +130,6 @@ steps: # Run on PRs even when the L4 step is optional (nightly only). optional: false label: ":amd: (MI355) DP EP Distributed NixlConnector PD accuracy" - automatic_shard: true dind: false device: mi355_4 timeout_in_minutes: 40 @@ -153,7 +145,6 @@ steps: - label: ":nvidia: (L4) [Nightly] CrossLayer KV layout Distributed NixlConnector PD accuracy" key: crosslayer-kv-layout-distributed-nixlconnector-pd-accuracy-tests-4-gpus - automatic_shard: true # Nightly only: 4-GPU L4 pods wait behind smaller jobs on per-PR runs. optional: true timeout_in_minutes: 55 @@ -172,7 +163,6 @@ steps: # Run on PRs even when the L4 step is optional (nightly only). optional: false label: ":amd: (MI300) CrossLayer KV layout Distributed NixlConnector PD accuracy" - automatic_shard: true dind: false device: mi300_4 timeout_in_minutes: 60 @@ -188,7 +178,6 @@ steps: - label: ":nvidia: (L4) [Nightly] Hybrid SSM NixlConnector PD accuracy" key: hybrid-ssm-nixlconnector-pd-accuracy-tests-4-gpus - automatic_shard: true # Nightly only: 4-GPU L4 pods wait behind smaller jobs on per-PR runs. optional: true timeout_in_minutes: 60 @@ -207,7 +196,6 @@ steps: # Run on PRs even when the L4 step is optional (nightly only). optional: false label: ":amd: (MI300) Hybrid SSM NixlConnector PD accuracy" - automatic_shard: true dind: false device: mi300_4 timeout_in_minutes: 80 @@ -229,7 +217,6 @@ steps: # configs keep one dedicated GPU per side. - label: ":nvidia: (L4) NixlConnector PD accuracy core" key: nixlconnector-pd-accuracy-core-2-gpus - automatic_shard: true timeout_in_minutes: 90 working_dir: "/vllm-workspace/tests" device: l4 @@ -249,7 +236,6 @@ steps: - label: ":nvidia: (L4) NixlConnector PD edge case" key: nixlconnector-pd-edge-cases-2-gpus - automatic_shard: true timeout_in_minutes: 40 working_dir: "/vllm-workspace/tests" device: l4 @@ -267,7 +253,6 @@ steps: mirror: amd: label: ":amd: (MI300) NixlConnector PD edge case" - automatic_shard: true dind: false device: mi300_2 timeout_in_minutes: 65 @@ -283,7 +268,6 @@ steps: - label: ":nvidia: (L4) Hybrid SSM NixlConnector PD prefix cache" key: hybrid-ssm-nixlconnector-pd-prefix-cache-2-gpus - automatic_shard: true timeout_in_minutes: 25 working_dir: "/vllm-workspace/tests" device: l4 @@ -299,7 +283,6 @@ steps: mirror: amd: label: ":amd: (MI355) Hybrid SSM NixlConnector PD prefix cache" - automatic_shard: true dind: false device: mi355_2 timeout_in_minutes: 25 @@ -316,7 +299,6 @@ steps: - label: ":nvidia: (L4) MultiConnector (Nixl+Offloading) PD accuracy" key: multiconnector-nixl-offloading-pd-accuracy-2-gpus - automatic_shard: true timeout_in_minutes: 40 working_dir: "/vllm-workspace/tests" device: l4 @@ -333,7 +315,6 @@ steps: mirror: amd: label: ":amd: (MI355) MultiConnector (Nixl+Offloading) PD accuracy" - automatic_shard: true dind: false device: mi355_2 timeout_in_minutes: 40 @@ -381,7 +362,6 @@ steps: - label: ":nvidia: (L4) MultiConnector (Nixl+Offloading) PD edge cases" key: multiconnector-nixl-offloading-pd-edge-cases-2-gpus - automatic_shard: true timeout_in_minutes: 25 working_dir: "/vllm-workspace/tests" device: l4 @@ -398,7 +378,6 @@ steps: mirror: amd: label: ":amd: (MI355) MultiConnector (Nixl+Offloading) PD edge cases" - automatic_shard: true dind: false device: mi355_2 timeout_in_minutes: 25 @@ -417,7 +396,6 @@ steps: # P TP 4 - D DPEP 4 test case for DSv4-Flash - label: ":nvidia: (H200) [Nightly] DSv4-Flash Disaggregated DP EP" key: dsv4-flash-disaggregated - automatic_shard: true timeout_in_minutes: 60 device: h200 optional: true @@ -443,7 +421,6 @@ steps: mirror: amd: label: ":amd: (MI355) DSv4-Flash Disaggregated DP EP" - automatic_shard: true dind: false device: mi355_8 timeout_in_minutes: 85 @@ -481,7 +458,6 @@ steps: - label: ":nvidia: (H200) [Nightly] Kimi-Linear-48B-A3B Disaggregated DP EP" key: kimi-linear-disaggregated - automatic_shard: true timeout_in_minutes: 60 device: h200 optional: true @@ -505,7 +481,6 @@ steps: mirror: amd: label: ":amd: (MI355) Kimi-Linear-48B-A3B Disaggregated DP EP" - automatic_shard: true dind: false device: mi355_4 num_devices: 4 diff --git a/.buildkite/test_areas/distributed.yaml b/.buildkite/test_areas/distributed.yaml index 9e1be206555c..28ecd3b3ae38 100644 --- a/.buildkite/test_areas/distributed.yaml +++ b/.buildkite/test_areas/distributed.yaml @@ -4,7 +4,6 @@ depends_on: steps: - label: ":nvidia: (L4) Distributed Comm Ops" key: distributed-comm-ops - automatic_shard: true timeout_in_minutes: 25 working_dir: "/vllm-workspace/tests" device: l4 @@ -24,7 +23,6 @@ steps: - pytest -v -s distributed/test_pipeline_partition.py mirror: amd: - automatic_shard: true label: ':amd: (MI300) Distributed Comm Ops' dind: false device: mi300_2 @@ -46,7 +44,6 @@ steps: - label: ":nvidia: (L4) Sharded RDT Weight Transfer" key: sharded-rdt-weight-transfer - automatic_shard: true device: l4 timeout_in_minutes: 20 working_dir: "/vllm-workspace/tests" @@ -62,7 +59,6 @@ steps: - pytest -v -s distributed/test_sharded_rdt_trainer.py mirror: amd: - automatic_shard: true label: ":amd: (MI355 DPX) Sharded RDT Weight Transfer" dind: false device: mi355_dpx @@ -87,7 +83,6 @@ steps: - label: ":nvidia: (L4) Distributed DP Basic" key: distributed-dp-tests-2-gpus - automatic_shard: true timeout_in_minutes: 45 working_dir: "/vllm-workspace/tests" device: l4 @@ -119,7 +114,6 @@ steps: amd: label: ":amd: (MI355) Distributed DP Basic" dind: false - automatic_shard: true device: mi355_2 timeout_in_minutes: 45 depends_on: @@ -138,7 +132,6 @@ steps: - label: ":nvidia: (L4) Distributed Compile + RPC" key: distributed-compile-rpc-tests-2-gpus - automatic_shard: true timeout_in_minutes: 65 working_dir: "/vllm-workspace/tests" device: l4 @@ -163,7 +156,6 @@ steps: - pytest -v -s ./compile/test_aot_compile.py::test_load_on_another_device mirror: amd: - automatic_shard: true label: ":amd: (MI355) Distributed Compile + RPC" dind: false device: mi355_2 @@ -188,7 +180,6 @@ steps: - label: ":nvidia: (L4) Distributed Torchrun + Shutdown" key: distributed-torchrun-shutdown-tests-2-gpus - automatic_shard: true timeout_in_minutes: 30 working_dir: "/vllm-workspace/tests" device: l4 @@ -213,7 +204,6 @@ steps: - pytest -v -s v1/worker/test_worker_memory_snapshot.py mirror: amd: - automatic_shard: true label: ':amd: (MI300) Distributed Torchrun + Shutdown' dind: false device: mi300_2 @@ -321,7 +311,6 @@ steps: key: distributed-dp-tests-4-gpus # Nightly only: 4-GPU L4 pods wait behind smaller jobs on per-PR runs. optional: true - automatic_shard: true timeout_in_minutes: 45 working_dir: "/vllm-workspace/tests" device: l4 @@ -353,7 +342,6 @@ steps: optional: false label: ':amd: (MI355) Distributed DP Extended' dind: false - automatic_shard: true device: mi355_4 timeout_in_minutes: 60 working_dir: /vllm-workspace/tests @@ -382,7 +370,6 @@ steps: - label: ":nvidia: (L4) [Nightly] Distributed Compile + Comm" key: distributed-compile-comm-4-gpus - automatic_shard: true # Nightly only: 4-GPU L4 pods wait behind smaller jobs on per-PR runs. optional: true timeout_in_minutes: 70 @@ -408,7 +395,6 @@ steps: - pytest -v -s distributed/test_multiproc_executor.py::test_multiproc_executor_multi_node mirror: amd: - automatic_shard: true # Run on PRs even when the L4 step is optional (nightly only). optional: false label: ':amd: (MI355) Distributed Compile + Comm' @@ -433,7 +419,6 @@ steps: - label: ":nvidia: (H100) [Nightly] Distributed DP + EP" key: distributed-tests-8xh100 - automatic_shard: true timeout_in_minutes: 20 optional: true device: h100 @@ -459,7 +444,6 @@ steps: - pytest -v -s distributed/test_eplb_execute.py -k symm_mem mirror: amd: - automatic_shard: true label: ":amd: (MI355) Distributed DP + EP" dind: false device: mi355_8 @@ -579,7 +563,6 @@ steps: - label: ":nvidia: (H100) [Nightly] Distributed Features" key: distributed-tests-2xh100-2xmi300 - automatic_shard: true timeout_in_minutes: 30 device: h100 optional: true @@ -596,7 +579,6 @@ steps: - pytest -v -s tests/distributed/test_packed_tensor.py mirror: amd: - automatic_shard: true label: ':amd: (MI355) Distributed Features' dind: false device: mi355_2 @@ -649,7 +631,6 @@ steps: - label: ":nvidia: (B200) [Nightly] Distributed" key: distributed-tests-2xb200 - automatic_shard: true device: b200-k8s optional: true working_dir: "/vllm-workspace/" @@ -663,7 +644,6 @@ steps: - pytest -v -s tests/distributed/test_pp_cudagraph.py::test_pp_cudagraph_mode_differs_across_stages mirror: amd: - automatic_shard: true label: ":amd: (MI355) Distributed AgRs All2All" dind: false device: mi355_2 @@ -686,7 +666,6 @@ steps: - label: ":nvidia: (B200) GEMM-RS/AR" key: gemm-rs-ar-2xb200 - automatic_shard: true timeout_in_minutes: 20 device: b200-k8s working_dir: "/vllm-workspace/" @@ -699,7 +678,6 @@ steps: - pytest -v -s tests/kernels/test_gemm_rs_ar.py mirror: amd: - automatic_shard: true label: ":amd: (MI355) GEMM-RS-AR" dind: false device: mi355_4 @@ -763,7 +741,6 @@ steps: key: pipeline-context-parallelism-4-gpus # Nightly only: 4-GPU L4 pods wait behind smaller jobs on per-PR runs. optional: true - automatic_shard: true timeout_in_minutes: 55 working_dir: "/vllm-workspace/tests" device: l4 @@ -794,7 +771,6 @@ steps: optional: false label: ":amd: (MI250) Pipeline + Context Parallelism" dind: false - automatic_shard: true device: mi250_4 timeout_in_minutes: 90 depends_on: @@ -819,7 +795,6 @@ steps: key: rayexecutorv2-4-gpus # Nightly only: 4-GPU L4 pods wait behind smaller jobs on per-PR runs. optional: true - automatic_shard: true timeout_in_minutes: 45 working_dir: "/vllm-workspace/tests" device: l4 @@ -844,7 +819,6 @@ steps: optional: false label: ':amd: (MI355) RayExecutorV2' dind: false - automatic_shard: true device: mi355_4 timeout_in_minutes: 75 working_dir: /vllm-workspace/tests @@ -867,7 +841,6 @@ steps: - label: ":nvidia: (L4) Distributed Preload" key: preload-2-gpus - automatic_shard: true timeout_in_minutes: 15 working_dir: "/vllm-workspace/tests" device: l4 @@ -881,7 +854,6 @@ steps: - VLLM_TRITON_FORCE_FIRST_CONFIG=1 pytest -v -s model_executor/model_loader/test_weight_cache.py -m 'distributed(num_gpus=2)' mirror: amd: - automatic_shard: true label: ":amd: (MI355) Distributed Preload" dind: false device: mi355_2 @@ -898,7 +870,6 @@ steps: - label: ":nvidia: (B200) Distributed Preload with Mooncake Store" key: preload-mooncake-store-2xb200 - automatic_shard: true timeout_in_minutes: 20 working_dir: "/vllm-workspace/tests" device: b200-k8s diff --git a/.buildkite/test_areas/engine.yaml b/.buildkite/test_areas/engine.yaml index 5e38895a545e..dff5d3ed58a7 100644 --- a/.buildkite/test_areas/engine.yaml +++ b/.buildkite/test_areas/engine.yaml @@ -4,7 +4,6 @@ depends_on: steps: - label: ":nvidia: (H200 MIG 18GB) Engine Core" key: engine - automatic_shard: true timeout_in_minutes: 30 device: h200_18gb source_file_dependencies: @@ -40,7 +39,6 @@ steps: amd: label: ":amd: (MI355 DPX) Engine Core" dind: false - automatic_shard: true device: mi355_dpx timeout_in_minutes: 40 depends_on: @@ -48,7 +46,6 @@ steps: - label: ":nvidia: (L4) V1 Engine" key: engine-1-gpu - automatic_shard: true device: l4 num_devices: 1 timeout_in_minutes: 45 @@ -65,7 +62,6 @@ steps: - pytest -v -s v1/test_tensor_ipc_queue.py mirror: amd: - automatic_shard: true label: ":amd: (MI250) V1 Engine" dind: false device: mi250_1 @@ -85,7 +81,6 @@ steps: - label: ":nvidia: (H200 MIG 18GB) E2E Scheduling" key: e2e-scheduling-1-gpu - automatic_shard: true timeout_in_minutes: 53 device: h200_18gb source_file_dependencies: @@ -96,7 +91,6 @@ steps: mirror: amd: label: ":amd: (MI250) E2E Scheduling" - automatic_shard: true dind: false device: mi250_1 timeout_in_minutes: 55 @@ -107,7 +101,6 @@ steps: - label: ":nvidia: (H200 MIG 18GB) E2E Scheduling Accuracy" key: e2e-scheduling-accuracy-1-gpu - automatic_shard: true timeout_in_minutes: 53 device: h200_18gb source_file_dependencies: @@ -118,7 +111,6 @@ steps: mirror: amd: label: ":amd: (MI250) E2E Scheduling Accuracy" - automatic_shard: true dind: false device: mi250_1 timeout_in_minutes: 55 @@ -130,7 +122,6 @@ steps: - label: ":nvidia: (H200 MIG 18GB) E2E Core" device: h200_18gb key: e2e-core-1-gpu - automatic_shard: true timeout_in_minutes: 65 source_file_dependencies: - vllm/v1/ @@ -145,7 +136,6 @@ steps: -k "not test_mamba_prefix_cache_mrv2" mirror: amd: - automatic_shard: true label: ":amd: (MI250) E2E Core" dind: false commands: @@ -169,7 +159,6 @@ steps: - label: ":nvidia: (H200 MIG 35GB) E2E Core Large Memory" device: h200_35gb key: e2e-core-large-memory - automatic_shard: true timeout_in_minutes: 50 source_file_dependencies: - vllm/v1/ @@ -183,7 +172,6 @@ steps: v1/e2e/general/test_mamba_prefix_cache.py::test_mamba_prefix_cache_mrv2_async_flashinfer mirror: amd: - automatic_shard: true label: ":amd: (MI355 DPX) E2E Core Large Memory" dind: false device: mi355_dpx @@ -207,7 +195,6 @@ steps: device: h100 num_devices: 1 key: replayssm-e2e - automatic_shard: true timeout_in_minutes: 90 working_dir: "/vllm-workspace/tests" source_file_dependencies: @@ -237,7 +224,6 @@ steps: - VLLM_TEST_REQUIRE_REPLAYSSM=1 VLLM_TEST_REPLAYSSM_NUM_GPUS=1 pytest -v kernels/mamba/test_ssu_dispatch.py::test_replayssm_materializes_only_accepted_boundary_with_compacted_rows kernels/mamba/test_ssu_dispatch.py::test_replayssm_postprocess_commits_staged_transition mirror: amd: - automatic_shard: true label: ":amd: (MI355) ReplaySSM E2E" dind: false device: mi355_1 @@ -305,7 +291,6 @@ steps: num_devices: 2 key: replayssm-e2e-nightly optional: true - automatic_shard: true timeout_in_minutes: 150 working_dir: "/vllm-workspace/tests" source_file_dependencies: @@ -334,7 +319,6 @@ steps: -k "not ((matches_baseline and not prefix_cache and not tp2) or decode_matches_baseline_tp2)" mirror: amd: - automatic_shard: true label: ":amd: (MI355) [Nightly] ReplaySSM E2E Prefix Cache and MTP" dind: false device: mi355_2 @@ -352,7 +336,6 @@ steps: - label: ":nvidia: (L4) [Nightly] V1 E2E" key: v1-e2e-2-gpus - automatic_shard: true timeout_in_minutes: 25 # TODO: Fix timeout after we have more confidence in the test stability optional: true device: l4 @@ -387,7 +370,6 @@ steps: v1/e2e/spec_decode/draft_model/test_structured_output_pp.py::test_specdec_pp_structured_output mirror: amd: - automatic_shard: true label: ":amd: (MI300) V1 E2E" dind: false device: mi300_2 @@ -397,7 +379,6 @@ steps: - label: ":nvidia: (B200) [Nightly] V1 E2E Hybrid Chunked Prefill" key: v1-e2e-hybrid-chunked-prefill-b200 - automatic_shard: true timeout_in_minutes: 35 device: b200-k8s num_devices: 1 @@ -410,7 +391,6 @@ steps: - pytest -v -s v1/e2e/test_hybrid_chunked_prefill.py mirror: amd: - automatic_shard: true label: ":amd: (MI355 DPX) V1 E2E Hybrid Chunked Prefill" dind: false device: mi355_dpx diff --git a/.buildkite/test_areas/entrypoints.yaml b/.buildkite/test_areas/entrypoints.yaml index 98abe7e8b828..349041b185eb 100644 --- a/.buildkite/test_areas/entrypoints.yaml +++ b/.buildkite/test_areas/entrypoints.yaml @@ -26,7 +26,6 @@ steps: - label: ":nvidia: (H200 MIG 35GB) Entrypoints Unit" device: h200_35gb key: entrypoints-unit-tests - automatic_shard: true timeout_in_minutes: 40 working_dir: "/vllm-workspace/tests" source_file_dependencies: @@ -43,7 +42,6 @@ steps: amd: label: ":amd: (MI355 DPX) Entrypoints Unit" dind: false - automatic_shard: true device: mi355_dpx timeout_in_minutes: 50 depends_on: @@ -54,7 +52,6 @@ steps: - label: ":nvidia: (H200 MIG 35GB) Entrypoints Integration (LLM)" device: h200_35gb key: entrypoints-integration-llm - automatic_shard: true timeout_in_minutes: 60 working_dir: "/vllm-workspace/tests" source_file_dependencies: @@ -74,7 +71,6 @@ steps: mirror: amd: label: ":amd: (MI355 DPX) Entrypoints Integration (LLM)" - automatic_shard: true dind: false device: mi355_dpx timeout_in_minutes: 90 @@ -113,7 +109,6 @@ steps: - label: ":nvidia: (H200 MIG 35GB) Entrypoints Integration (OpenAI API completion)" device: h200_35gb key: entrypoints-integration-api-server-openai-completion - automatic_shard: true timeout_in_minutes: 68 working_dir: "/vllm-workspace/tests" source_file_dependencies: @@ -136,7 +131,6 @@ steps: mirror: amd: label: ":amd: (MI355 DPX) Entrypoints Integration (OpenAI API completion)" - automatic_shard: true dind: false device: mi355_dpx timeout_in_minutes: 65 @@ -146,7 +140,6 @@ steps: - label: ":nvidia: (H200 MIG 35GB) Entrypoints Integration (OpenAI API chat_completion)" device: h200_35gb key: entrypoints-integration-api-server-openai-chat_completion - automatic_shard: true timeout_in_minutes: 83 working_dir: "/vllm-workspace/tests" source_file_dependencies: @@ -158,7 +151,6 @@ steps: - pytest -v -s entrypoints/openai/chat_completion mirror: amd: - automatic_shard: true label: ":amd: (MI355 DPX) Entrypoints Integration (OpenAI API chat_completion)" dind: false device: mi355_dpx @@ -169,7 +161,6 @@ steps: - label: ":nvidia: (H200 MIG 35GB) Entrypoints Integration (API Server Generate)" device: h200_35gb key: entrypoints-integration-api-server-generate - automatic_shard: true timeout_in_minutes: 50 working_dir: "/vllm-workspace/tests" source_file_dependencies: @@ -188,7 +179,6 @@ steps: - pytest -v -s entrypoints/cohere mirror: amd: - automatic_shard: true label: ":amd: (MI355 DPX) Entrypoints Integration (API Server Generate)" dind: false device: mi355_dpx @@ -199,7 +189,6 @@ steps: - label: ":nvidia: (H200 MIG 35GB) Entrypoints Integration (Responses API)" device: h200_35gb key: entrypoints-integration-responses-api - automatic_shard: true timeout_in_minutes: 50 working_dir: "/vllm-workspace/tests" source_file_dependencies: @@ -211,7 +200,6 @@ steps: mirror: amd: label: ":amd: (MI355 DPX) Entrypoints Integration (Responses API)" - automatic_shard: true dind: false device: mi355_dpx timeout_in_minutes: 50 @@ -221,7 +209,6 @@ steps: - label: ":nvidia: (H200 MIG 35GB) Entrypoints Integration (Speech to Text)" device: h200_35gb key: entrypoints-integration-speech_to_text - automatic_shard: true timeout_in_minutes: 45 working_dir: "/vllm-workspace/tests" source_file_dependencies: @@ -235,7 +222,6 @@ steps: amd: label: ":amd: (MI355 DPX) Entrypoints Integration (Speech to Text)" dind: false - automatic_shard: true device: mi355_dpx timeout_in_minutes: 60 working_dir: "/vllm-workspace/tests" @@ -245,7 +231,6 @@ steps: - label: ":nvidia: (H200 MIG 35GB) Entrypoints Integration (Multimodal)" device: h200_35gb key: entrypoints-integration-multimodal - automatic_shard: true timeout_in_minutes: 45 working_dir: "/vllm-workspace/tests" source_file_dependencies: @@ -259,7 +244,6 @@ steps: amd: label: ":amd: (MI355 DPX) Entrypoints Integration (Multimodal)" dind: false - automatic_shard: true device: mi355_dpx timeout_in_minutes: 55 depends_on: @@ -268,7 +252,6 @@ steps: - label: ":nvidia: (H200 MIG 35GB) Entrypoints Integration (Pooling)" device: h200_35gb key: entrypoints-integration-pooling - automatic_shard: true timeout_in_minutes: 75 working_dir: "/vllm-workspace/tests" source_file_dependencies: @@ -282,7 +265,6 @@ steps: amd: label: ":amd: (MI355 DPX) Entrypoints Integration (Pooling)" dind: false - automatic_shard: true device: mi355_dpx timeout_in_minutes: 65 working_dir: "/vllm-workspace/tests" @@ -291,7 +273,6 @@ steps: - label: ":nvidia: (H200 MIG 18GB) OpenAI API Correctness" key: openai-api-correctness - automatic_shard: true timeout_in_minutes: 20 device: h200_18gb source_file_dependencies: @@ -301,7 +282,6 @@ steps: - pytest -s entrypoints/openai/correctness/ mirror: amd: - automatic_shard: true label: ":amd: (MI355 DPX) OpenAI API Correctness" dind: false device: mi355_dpx diff --git a/.buildkite/test_areas/expert_parallelism.yaml b/.buildkite/test_areas/expert_parallelism.yaml index 0266f244d9cd..74381631f767 100644 --- a/.buildkite/test_areas/expert_parallelism.yaml +++ b/.buildkite/test_areas/expert_parallelism.yaml @@ -4,7 +4,6 @@ depends_on: steps: - label: ":nvidia: (H200 MIG 18GB) EPLB Algorithm" key: eplb-algorithm - automatic_shard: true timeout_in_minutes: 20 device: h200_18gb working_dir: "/vllm-workspace/tests" @@ -19,7 +18,6 @@ steps: - pytest -v -s distributed/test_eplb_utils.py mirror: amd: - automatic_shard: true label: ":amd: (MI355 DPX) EPLB Algorithm" dind: false device: mi355_dpx @@ -34,7 +32,6 @@ steps: - label: ":nvidia: (L4) [Nightly] EPLB Execution" key: eplb-execution - automatic_shard: true # Nightly only: 4-GPU L4 pods wait behind smaller jobs on per-PR runs. optional: true timeout_in_minutes: 25 @@ -49,7 +46,6 @@ steps: - pytest -v -s distributed/test_eplb_spec_decode.py mirror: amd: - automatic_shard: true # Run on PRs even when the L4 step is optional (nightly only). optional: false label: ":amd: (MI355) EPLB Execution" @@ -66,7 +62,6 @@ steps: - label: ":nvidia: (B200) [Nightly] EP Weight Filter with EPLB" key: ep-weight-filter-eplb-2xb200 - automatic_shard: true optional: true timeout_in_minutes: 20 working_dir: "/vllm-workspace/tests" @@ -82,7 +77,6 @@ steps: mirror: amd: label: ":amd: (MI355) EP Weight Filter with EPLB" - automatic_shard: true optional: false dind: false device: mi355_2 diff --git a/.buildkite/test_areas/fault_tolerance.yaml b/.buildkite/test_areas/fault_tolerance.yaml index a8b525cbf59a..ce830c42862d 100644 --- a/.buildkite/test_areas/fault_tolerance.yaml +++ b/.buildkite/test_areas/fault_tolerance.yaml @@ -4,7 +4,6 @@ depends_on: steps: - label: ":nvidia: (H100) Fault Tolerance E2E" key: fault-tolerance-e2e-4xh100 - automatic_shard: true timeout_in_minutes: 35 device: h100 num_devices: 4 @@ -31,7 +30,6 @@ steps: mirror: amd: label: ":amd: (MI355) MoRI EP Healthy Baseline" - automatic_shard: true dind: false device: mi355_2 num_devices: 2 diff --git a/.buildkite/test_areas/jit_monitor.yaml b/.buildkite/test_areas/jit_monitor.yaml index 2e606a988ca0..9045f8f8306c 100644 --- a/.buildkite/test_areas/jit_monitor.yaml +++ b/.buildkite/test_areas/jit_monitor.yaml @@ -4,7 +4,6 @@ depends_on: steps: - label: ":nvidia: (H200 MIG 35GB) No Runtime JITs E2E" key: jit-monitor-no-runtime-jit - automatic_shard: true device: h200_35gb timeout_in_minutes: 45 source_file_dependencies: @@ -23,7 +22,6 @@ steps: - pytest -v -s jit_monitor/test_no_runtime_jit.py --timeout=900 --timeout-method=thread mirror: amd: - automatic_shard: true # Focused MRV2 regression; the broader model/backend matrix is pending. label: ':amd: (MI355 DPX) MRV2 Sampler JIT Warmup' dind: false diff --git a/.buildkite/test_areas/kernels.yaml b/.buildkite/test_areas/kernels.yaml index e6feaf965246..a01dd87f5b69 100644 --- a/.buildkite/test_areas/kernels.yaml +++ b/.buildkite/test_areas/kernels.yaml @@ -4,7 +4,6 @@ depends_on: steps: - label: ":nvidia: (H200 MIG 18GB) vLLM IR" key: vllm-ir-tests - automatic_shard: true timeout_in_minutes: 35 device: h200_18gb working_dir: "/vllm-workspace/" @@ -16,7 +15,6 @@ steps: - pytest -v -s tests/kernels/ir mirror: amd: - automatic_shard: true label: ":amd: (MI355 DPX) vLLM IR" dind: false device: mi355_dpx @@ -35,7 +33,6 @@ steps: - label: ":nvidia: (H200 MIG 35GB) Core Operation Kernels" device: h200_35gb key: kernels-core-operation-test - automatic_shard: true timeout_in_minutes: 120 source_file_dependencies: - csrc/ @@ -55,7 +52,6 @@ steps: amd: label: ":amd: (MI300) Core Operation Kernels" dind: false - automatic_shard: true device: mi300_1 timeout_in_minutes: 90 working_dir: "/vllm-workspace/tests" @@ -89,7 +85,6 @@ steps: - label: ":nvidia: (H100) MiniMax Reduce RMS Kernels" key: kernels-minimax-reduce-rms-test-2-gpus - automatic_shard: true timeout_in_minutes: 20 num_devices: 2 device: h100 @@ -105,7 +100,6 @@ steps: - pytest -v -s kernels/core/test_minimax_reduce_rms.py mirror: amd: - automatic_shard: true label: ":amd: (MI355) MiniMax Reduce RMS Kernels" dind: false device: mi355_2 @@ -128,7 +122,6 @@ steps: - label: ":nvidia: (H100) DeepSeek V4 Kernels" key: deepseek-v4-kernel-test-h100 - automatic_shard: true timeout_in_minutes: 30 device: h100 source_file_dependencies: @@ -143,7 +136,6 @@ steps: amd: label: ":amd: (MI355 DPX) DeepSeek V4 Fused Kernels" dind: false - automatic_shard: true device: mi355_dpx num_devices: 1 timeout_in_minutes: 60 @@ -169,7 +161,6 @@ steps: - label: ":nvidia: (H100) BF16x3 Router GEMM Kernels" key: kernels-bf16x3-router-gemm-h100 - automatic_shard: true timeout_in_minutes: 20 device: h100 source_file_dependencies: @@ -180,7 +171,6 @@ steps: - pytest -v -s kernels/test_bf16x3_router_gemm_cutedsl.py mirror: amd: - automatic_shard: true label: ":amd: (MI355) BF16x3 Router GEMM Kernels" dind: false device: mi355_1 @@ -197,7 +187,6 @@ steps: - label: ":nvidia: (B200) DeepSeek V4 Kernels" key: deepseek-v4-kernel-test-b200 - automatic_shard: true timeout_in_minutes: 20 device: b200-k8s source_file_dependencies: @@ -221,7 +210,6 @@ steps: - pytest -v -s model_executor/layers/test_mla_short_prefill_indexer.py -k sparse_mqa_logits mirror: amd: - automatic_shard: true label: ":amd: (MI355) DeepSeek V4 MoE Integration" dind: false device: mi355_1 @@ -254,7 +242,6 @@ steps: # ptxas override. - label: ":nvidia: (B200) mHC Kernels" key: kernels-mhc-test-b200 - automatic_shard: true timeout_in_minutes: 25 device: b200-k8s source_file_dependencies: @@ -266,7 +253,6 @@ steps: - pytest -v -s kernels/mhc mirror: amd: - automatic_shard: true label: ":amd: (MI355) mHC Kernels" dind: false device: mi355_1 @@ -282,7 +268,6 @@ steps: - label: ":nvidia: (B200) FLA Ops Kernels" key: kernels-fla-ops-test-b200 - automatic_shard: true timeout_in_minutes: 25 device: b200-k8s source_file_dependencies: @@ -294,7 +279,6 @@ steps: - pytest -v -s kernels/fla mirror: amd: - automatic_shard: true label: ":amd: (MI355) FLA Ops Kernels" dind: false device: mi355_1 @@ -310,7 +294,6 @@ steps: - label: ":nvidia: (B200) MiniMax M3 MegaMoE" key: minimax-m3-mega-moe-test-b200 - automatic_shard: true timeout_in_minutes: 20 device: b200-k8s source_file_dependencies: @@ -323,7 +306,6 @@ steps: - label: ":nvidia: (B200) MiniMax-M3 Model Tests" key: minimax-m3-model-tests-b200 - automatic_shard: true timeout_in_minutes: 20 device: b200-k8s source_file_dependencies: @@ -336,7 +318,6 @@ steps: mirror: amd: label: ":amd: (MI355) MiniMax-M3 Model Tests" - automatic_shard: true dind: false device: mi355_1 timeout_in_minutes: 20 @@ -358,7 +339,6 @@ steps: # CuTeDSL Kernels (B200)). - label: ":nvidia: (B200) Miscellaneous Kernels" key: kernels-root-misc-test-b200 - automatic_shard: true timeout_in_minutes: 30 device: b200-k8s source_file_dependencies: @@ -392,7 +372,6 @@ steps: mirror: amd: label: ":amd: (MI355) Miscellaneous Kernels" - automatic_shard: true dind: false device: mi355_1 timeout_in_minutes: 55 @@ -437,7 +416,6 @@ steps: - label: ":nvidia: (L4) Attention Kernels" key: kernels-attention-test - automatic_shard: true device: l4 num_devices: 1 timeout_in_minutes: 30 @@ -454,7 +432,6 @@ steps: mirror: amd: label: ":amd: (MI300) Attention Kernels" - automatic_shard: true dind: false device: mi300_1 timeout_in_minutes: 90 @@ -486,7 +463,6 @@ steps: - label: ":nvidia: (H100) Attention DiffKV Kernels" key: kernels-attention-diffkv-test-h100 - automatic_shard: true timeout_in_minutes: 20 device: h100 num_devices: 1 @@ -499,7 +475,6 @@ steps: - pytest -v -s kernels/attention/test_triton_unified_attention_diffkv.py mirror: amd: - automatic_shard: true label: ":amd: (MI355 DPX) Attention DiffKV Kernels" dind: false device: mi355_dpx @@ -519,7 +494,6 @@ steps: - label: ":nvidia: (H100) MLA Kernel Test" key: kernels-flashmla-test-h100 - automatic_shard: true timeout_in_minutes: 25 device: h100 num_devices: 1 @@ -541,7 +515,6 @@ steps: - pytest -v -s kernels/attention/test_mla_cross_layer_kernel_equivalence.py mirror: amd: - automatic_shard: true label: ":amd: (MI355 DPX) MLA Kernels" dind: false device: mi355_dpx @@ -592,7 +565,6 @@ steps: - label: ":nvidia: (L4) Quantization Kernels" key: kernels-quantization-test - automatic_shard: true device: l4 num_devices: 1 timeout_in_minutes: 30 @@ -609,7 +581,6 @@ steps: - pytest -v -s kernels/quantization/test_fp8_quant.py mirror: amd: - automatic_shard: true label: ":amd: (MI300 DPX) Quantization Kernels" dind: false device: mi300_dpx @@ -637,7 +608,6 @@ steps: - label: ":nvidia: (L4) MoE Kernels" key: kernels-moe-test - automatic_shard: true device: l4 num_devices: 1 timeout_in_minutes: 50 @@ -658,7 +628,6 @@ steps: mirror: amd: label: ":amd: (MI300) MoE Kernels" - automatic_shard: true dind: false device: mi300_1 timeout_in_minutes: 55 @@ -684,7 +653,6 @@ steps: - label: ":nvidia: (H200 MIG 35GB) Mamba Kernels" device: h200_35gb key: kernels-mamba-test - automatic_shard: true timeout_in_minutes: 60 source_file_dependencies: - csrc/mamba/ @@ -700,7 +668,6 @@ steps: amd: label: ":amd: (MI250) Mamba Kernels" dind: false - automatic_shard: true device: mi250_1 timeout_in_minutes: 55 depends_on: @@ -711,7 +678,6 @@ steps: - label: ":nvidia: (H100) DeepGEMM Kernels" key: kernels-deepgemm-test-h100 - automatic_shard: true timeout_in_minutes: 35 device: h100 num_devices: 1 @@ -737,7 +703,6 @@ steps: - pytest -v -s quantization/test_cutlass_w4a16.py mirror: amd: - automatic_shard: true label: ":amd: (MI355) AITER FP8 GEMM and MoE, MQA Logits and W4A16 Kernels" dind: false device: mi355_1 @@ -775,7 +740,6 @@ steps: - label: ":nvidia: (B200) MiniMax-M3 Attention Kernels" key: kernels-minimax-m3-attention-b200 - automatic_shard: true timeout_in_minutes: 30 working_dir: "/vllm-workspace/" device: b200-k8s @@ -790,7 +754,6 @@ steps: - pytest -v -s tests/kernels/attention/test_minimax_m3.py mirror: amd: - automatic_shard: true label: ":amd: (MI355 DPX) MiniMax-M3 Attention Kernels" dind: false device: mi355_dpx @@ -810,7 +773,6 @@ steps: - label: ":nvidia: (H100) MiniMax-M3 Sparse Prefill Kernels" key: kernels-minimax-m3-sparse-prefill-h100 - automatic_shard: true timeout_in_minutes: 15 working_dir: "/vllm-workspace/" device: h100 @@ -828,7 +790,6 @@ steps: - label: ":nvidia: (B200) Kernels" key: kernels-b200 - automatic_shard: true timeout_in_minutes: 30 working_dir: "/vllm-workspace/" device: b200-k8s @@ -912,7 +873,6 @@ steps: - pytest -v -s tests/kernels/mamba/test_gdn_prefill_cutedsl.py mirror: amd: - automatic_shard: true label: ":amd: (MI355) NVFP4 Dense Emulation" dind: false device: mi355_1 @@ -940,7 +900,6 @@ steps: - label: ":nvidia: (B200) Gate Linear CuTeDSL Kernels" key: kernels-gate-linear-cutedsl-b200 - automatic_shard: true timeout_in_minutes: 15 working_dir: "/vllm-workspace/" device: b200-k8s @@ -962,7 +921,6 @@ steps: - pytest -v -s tests/kernels/test_ll_fp32w_gemm.py mirror: amd: - automatic_shard: true label: ":amd: (MI355) Gate Linear Router Kernels" dind: false device: mi355_1 @@ -1049,7 +1007,6 @@ steps: - label: ":nvidia: (H100) [Nightly] FP8 MoE Kernels" key: kernels-fp8-moe-test-1xh100 - automatic_shard: true timeout_in_minutes: 40 device: h100 num_devices: 1 @@ -1067,7 +1024,6 @@ steps: - label: ":nvidia: (H100) [Nightly] DeepEP FP8 MoE Kernels" key: kernels-fp8-moe-test-2xh100 - automatic_shard: true timeout_in_minutes: 45 device: h100 num_devices: 2 @@ -1078,7 +1034,6 @@ steps: mirror: amd: label: ":amd: (MI355) DeepEP FP8 MoE Kernels" - automatic_shard: true dind: false device: mi355_2 timeout_in_minutes: 55 @@ -1098,7 +1053,6 @@ steps: - label: ":nvidia: (B200) [Nightly] FP4 MoE Kernels" key: kernels-fp4-moe-test-b200 - automatic_shard: true timeout_in_minutes: 25 device: b200-k8s num_devices: 1 @@ -1110,7 +1064,6 @@ steps: - pytest -v -s kernels/moe/test_ocp_mx_moe.py mirror: amd: - automatic_shard: true label: ":amd: (MI355 DPX) FP4 and INT4 MoE Kernels" dind: false device: mi355_dpx @@ -1139,7 +1092,6 @@ steps: - label: ":nvidia: (H100) FusedMoE Layer Kernels" key: kernels-fusedmoe-layer-test-2-h100s - automatic_shard: true timeout_in_minutes: 30 device: h100 num_devices: 2 @@ -1157,7 +1109,6 @@ steps: amd: label: ":amd: (MI355) FusedMoE Layer Kernels" dind: false - automatic_shard: true device: mi355_2 timeout_in_minutes: 35 working_dir: "/vllm-workspace/tests" @@ -1184,7 +1135,6 @@ steps: - label: ":nvidia: (B200) FusedMoE Layer Kernels" key: kernels-fusedmoe-layer-test-2-b200s - automatic_shard: true timeout_in_minutes: 45 device: b200-k8s num_devices: 2 @@ -1201,7 +1151,6 @@ steps: - pytest -v -s kernels/moe/test_deepep_v2_moe.py mirror: amd: - automatic_shard: true label: ":amd: (MI355) MoRI EP Numerics" dind: false device: mi355_2 @@ -1235,7 +1184,6 @@ steps: - label: ":amd: (MI355 DPX) Native Quantization Kernels" key: amd-native-quantization-kernels-mi355 - automatic_shard: true dind: false device: mi355_dpx num_devices: 1 @@ -1264,7 +1212,6 @@ steps: - label: ":amd: (MI355 DPX) FP8 MoE Kernels" key: amd-fp8-moe-kernels-mi355 - automatic_shard: true dind: false device: mi355_dpx num_devices: 1 @@ -1295,7 +1242,6 @@ steps: - label: ":amd: (MI355 DPX) Kernels" key: amd-kernels-mi355 - automatic_shard: true dind: false device: mi355_dpx num_devices: 1 @@ -1326,7 +1272,6 @@ steps: - label: ":amd: (MI355) Mamba SSM Kernels" key: amd-mamba-ssm-kernels-mi355 - automatic_shard: true dind: false device: mi355_1 num_devices: 1 diff --git a/.buildkite/test_areas/lm_eval.yaml b/.buildkite/test_areas/lm_eval.yaml index 9f8bb833eb82..9c549d043b0a 100644 --- a/.buildkite/test_areas/lm_eval.yaml +++ b/.buildkite/test_areas/lm_eval.yaml @@ -5,7 +5,6 @@ steps: - label: ":nvidia: (H200 MIG 35GB) LM Eval Small Models" device: h200_35gb key: lm-eval-small-models - automatic_shard: true timeout_in_minutes: 45 source_file_dependencies: - csrc/ @@ -15,7 +14,6 @@ steps: - pytest -s -v evals/gsm8k/test_gsm8k_correctness.py --config-list-file=configs/models-small.yaml mirror: amd: - automatic_shard: true label: ":amd: (MI300 DPX) LM Eval Small Models" dind: false device: mi300_dpx @@ -34,7 +32,6 @@ steps: - label: ":amd: (MI300 DPX) LM Eval Small Models Harness" key: amd-lm-eval-small-models-harness - automatic_shard: true dind: false device: mi300_dpx timeout_in_minutes: 35 @@ -57,7 +54,6 @@ steps: - label: ":nvidia: (H200 MIG 35GB) LM Eval Watermarking" device: h200_35gb key: lm-eval-watermarking - automatic_shard: true # Recorded runs took 80–140 seconds timeout_in_minutes: 10 source_file_dependencies: @@ -77,7 +73,6 @@ steps: - pytest -s -v watermarking/test_watermarking_e2e.py mirror: amd: - automatic_shard: true label: ":amd: (MI355 DPX) LM Eval Watermarking" dind: false device: mi355_dpx @@ -101,7 +96,6 @@ steps: - label: ":nvidia: (H100) [Nightly] LM Eval Large Models" key: lm-eval-large-models-4xh100 - automatic_shard: true device: h100 optional: true num_devices: 4 @@ -115,7 +109,6 @@ steps: mirror: amd: label: ":amd: (MI355) LM Eval Large Models FP8" - automatic_shard: true dind: false device: mi355_4 timeout_in_minutes: 45 @@ -137,7 +130,6 @@ steps: - label: ":nvidia: (B200) [Nightly] LM Eval Small Models" key: lm-eval-small-models-1xb200 - automatic_shard: true timeout_in_minutes: 50 device: b200-k8s optional: true @@ -149,7 +141,6 @@ steps: mirror: amd: label: ":amd: (MI355 DPX) LM Eval Small Models" - automatic_shard: true dind: false device: mi355_dpx num_devices: 1 @@ -175,7 +166,6 @@ steps: - label: ":nvidia: (B200) [Nightly] LM Eval Small Models Distributed" key: lm-eval-small-models-distributed-2xb200 - automatic_shard: true timeout_in_minutes: 120 device: b200-k8s num_devices: 2 @@ -189,7 +179,6 @@ steps: mirror: amd: label: ":amd: (MI355) LM Eval Small Models Distributed" - automatic_shard: true dind: false device: mi355_2 timeout_in_minutes: 145 @@ -218,7 +207,6 @@ steps: - label: ":nvidia: (B200) [Nightly] LM Eval HiSparse" key: lm-eval-hisparse-nightly-4xb200 - automatic_shard: true timeout_in_minutes: 60 device: b200-k8s num_devices: 4 @@ -237,7 +225,6 @@ steps: - label: ":nvidia: (B200) [Nightly] LM Eval PCP" key: lm-eval-pcp-4xb200 - automatic_shard: true timeout_in_minutes: 360 device: b200-k8s num_devices: 4 @@ -265,7 +252,6 @@ steps: mirror: amd: label: ":amd: (MI355) LM Eval PCP" - automatic_shard: true dind: false device: mi355_4 num_devices: 4 @@ -299,7 +285,6 @@ steps: - label: ":nvidia: (H100) LM Eval Watermark Feature Combination" key: lm-eval-dspark-watermark-2xh100 - automatic_shard: true device: h100 num_devices: 2 # Local TP2 runs took 84–379 seconds, including startup. @@ -331,7 +316,6 @@ steps: watermarking/test_watermarking_e2e.py mirror: amd: - automatic_shard: true label: ":amd: (MI355) LM Eval Watermark Feature Combination" dind: false device: mi355_2 @@ -490,7 +474,6 @@ steps: - label: ":nvidia: (B200) [Nightly] LM Eval Large Models EP" key: lm-eval-large-models-ep-2xb200 - automatic_shard: true timeout_in_minutes: 60 device: b200-k8s optional: true @@ -503,7 +486,6 @@ steps: mirror: amd: label: ":amd: (MI355) LM Eval Large Models EP" - automatic_shard: true dind: false device: mi355_2 num_devices: 2 @@ -545,7 +527,6 @@ steps: - label: ":nvidia: (B200) [Nightly] LM Eval Qwen3.5 Models" key: lm-eval-qwen3-5-models-2xb200 - automatic_shard: true timeout_in_minutes: 45 device: b200-k8s optional: true @@ -562,7 +543,6 @@ steps: mirror: amd: label: ":amd: (MI355) LM Eval Qwen3.5 Models" - automatic_shard: true dind: false device: mi355_2 timeout_in_minutes: 75 @@ -597,7 +577,6 @@ steps: - label: ":nvidia: (H200) [Nightly] LM Eval DeepSeek R1 TP" key: lm-eval-deepseek-r1-tp-8xh200 - automatic_shard: true timeout_in_minutes: 35 device: h200 optional: true @@ -608,7 +587,6 @@ steps: amd: label: ":amd: (MI355) LM Eval DeepSeek R1 TP" dind: false - automatic_shard: true device: mi355_4 num_devices: 4 timeout_in_minutes: 65 @@ -643,7 +621,6 @@ steps: - label: ":nvidia: (H200) [Nightly] LM Eval DeepSeek R1 DP" key: lm-eval-deepseek-r1-dp-8xh200 - automatic_shard: true timeout_in_minutes: 35 device: h200 optional: true @@ -653,7 +630,6 @@ steps: mirror: amd: label: ":amd: (MI355) LM Eval DeepSeek R1 DP" - automatic_shard: true dind: false device: mi355_4 num_devices: 4 @@ -689,7 +665,6 @@ steps: - label: ":nvidia: (H200) [Nightly] LM Eval DeepSeek V3.2 TP" key: lm-eval-deepseek-v32-tp-8xh200 - automatic_shard: true timeout_in_minutes: 35 device: h200 optional: true @@ -700,7 +675,6 @@ steps: amd: label: ":amd: (MI355) LM Eval DeepSeek V3-2 TP" dind: false - automatic_shard: true device: mi355_4 num_devices: 4 timeout_in_minutes: 65 @@ -735,7 +709,6 @@ steps: - label: ":nvidia: (H200) [Nightly] LM Eval DeepSeek V3.2 DP" key: lm-eval-deepseek-v32-dp-8xh200 - automatic_shard: true timeout_in_minutes: 35 device: h200 optional: true @@ -745,7 +718,6 @@ steps: mirror: amd: label: ":amd: (MI355) LM Eval DeepSeek V3-2 DP" - automatic_shard: true dind: false device: mi355_4 num_devices: 4 @@ -781,7 +753,6 @@ steps: - label: ":nvidia: (H200) [Nightly] LM Eval Nemotron 3 Super" key: lm-eval-nemotron-3-super-8xh200 - automatic_shard: true timeout_in_minutes: 35 device: h200 optional: true @@ -790,7 +761,6 @@ steps: - pytest -s -v evals/gsm8k/test_gsm8k_correctness.py --config-list-file=configs/models-h200-nemotron-3-super.txt mirror: amd: - automatic_shard: true label: ":amd: (MI355) LM Eval Nemotron 3 Super" dind: false device: mi355_4 @@ -827,7 +797,6 @@ steps: - label: ":amd: (MI355) [Nightly] LM Eval Large Models" key: amd-lm-eval-large-models - automatic_shard: true dind: false device: mi355_8 optional: true @@ -843,7 +812,6 @@ steps: - label: ":nvidia: (H100) [Nightly] MoE Refactor Integration TEMPORARY" key: moe-refactor-integration-test-h100-temporary - automatic_shard: true device: h100 optional: true num_devices: 2 @@ -851,7 +819,6 @@ steps: - pytest -s -v evals/gsm8k/test_gsm8k_correctness.py --config-list-file=evals/gsm8k/configs/moe-refactor/config-h100.txt mirror: amd: - automatic_shard: true label: ":amd: (MI355) MoE Refactor Integration" dind: false device: mi355_2 @@ -873,7 +840,6 @@ steps: - label: ":nvidia: (B200) [Nightly] MoE Refactor Integration TEMPORARY" key: moe-refactor-integration-test-b200-temporary - automatic_shard: true device: b200-k8s optional: true num_devices: 2 @@ -881,7 +847,6 @@ steps: - pytest -s -v evals/gsm8k/test_gsm8k_correctness.py --config-list-file=evals/gsm8k/configs/moe-refactor/config-b200.txt mirror: amd: - automatic_shard: true label: ":amd: (MI355) MoE Refactor BF16 Integration" dind: false device: mi355_2 @@ -908,7 +873,6 @@ steps: - label: ":nvidia: (B200) [Nightly] MoE Refactor DP Integration TEMPORARY" key: moe-refactor-integration-test-b200-dp-temporary - automatic_shard: true device: b200-k8s optional: true num_devices: 2 @@ -916,7 +880,6 @@ steps: - pytest -s -v evals/gsm8k/test_gsm8k_correctness.py --config-list-file=evals/gsm8k/configs/moe-refactor-dp-ep/config-b200.txt mirror: amd: - automatic_shard: true label: ":amd: (MI355) MoE Refactor BF16 DP Integration" dind: false device: mi355_2 @@ -944,7 +907,6 @@ steps: - label: ":nvidia: (B200) MoE Refactor DP Integration (DeepEP v2)" key: moe-refactor-dp-ep-test-b200-deepep-v2 - automatic_shard: true device: b200-k8s # DeepEP v2 needs every rank on one NVLink node, and EP_DISABLE_GIN drops its # NCCL GIN/RDMA requirement at num_ranks > 1. @@ -968,7 +930,6 @@ steps: - label: ":nvidia: (L4) [Nightly] LM Eval Humming TEMPORARY" key: lm-eval-humming-l4 - automatic_shard: true optional: true timeout_in_minutes: 90 device: l4 @@ -988,7 +949,6 @@ steps: - label: ":nvidia: (H100) [Nightly] LM Eval Humming Default TEMPORARY" key: lm-eval-humming-h100-default - automatic_shard: true optional: true timeout_in_minutes: 70 device: h100 @@ -1004,7 +964,6 @@ steps: - label: ":nvidia: (H100) [Nightly] LM Eval Humming Activation FP8/INT8 TEMPORARY" key: lm-eval-humming-h100-act - automatic_shard: true optional: true timeout_in_minutes: 90 device: h100 @@ -1021,7 +980,6 @@ steps: - label: ":nvidia: (H100) [Nightly] LM Eval Humming DeepEP TEMPORARY" key: lm-eval-humming-h100-deepep - automatic_shard: true optional: true timeout_in_minutes: 60 device: h100 @@ -1042,7 +1000,6 @@ steps: - label: ":nvidia: (B200) [Nightly] LM Eval Humming Default TEMPORARY" key: lm-eval-humming-b200-default - automatic_shard: true optional: true timeout_in_minutes: 50 device: b200-k8s @@ -1058,7 +1015,6 @@ steps: - label: ":nvidia: (B200) [Nightly] LM Eval Humming Activation FP8/INT8 TEMPORARY" key: lm-eval-humming-b200-act - automatic_shard: true optional: true timeout_in_minutes: 50 device: b200-k8s @@ -1075,7 +1031,6 @@ steps: - label: ":nvidia: (H200 MIG 18GB) LM Eval TurboQuant k8v4" key: lm-eval-turboquant-k8v4 - automatic_shard: true timeout_in_minutes: 30 device: h200_18gb source_file_dependencies: @@ -1087,7 +1042,6 @@ steps: - pytest -s -v evals/gsm8k/test_gsm8k_correctness.py --config-list-file=evals/gsm8k/configs/models-turboquant-k8v4.txt mirror: amd: - automatic_shard: true label: ":amd: (MI355 DPX) LM Eval TurboQuant k8v4" dind: false device: mi355_dpx @@ -1117,7 +1071,6 @@ steps: - label: ":nvidia: (H200 MIG 18GB) LM Eval TurboQuant t4nc" key: lm-eval-turboquant-t4nc - automatic_shard: true timeout_in_minutes: 30 device: h200_18gb source_file_dependencies: @@ -1131,7 +1084,6 @@ steps: amd: label: ":amd: (MI355 DPX) LM Eval TurboQuant t4nc" dind: false - automatic_shard: true device: mi355_dpx num_devices: 1 timeout_in_minutes: 30 @@ -1159,7 +1111,6 @@ steps: - label: ":nvidia: (H200 MIG 18GB) LM Eval TurboQuant k3v4nc" key: lm-eval-turboquant-k3v4nc - automatic_shard: true timeout_in_minutes: 30 device: h200_18gb source_file_dependencies: @@ -1173,7 +1124,6 @@ steps: amd: label: ":amd: (MI355 DPX) LM Eval TurboQuant k3v4nc" dind: false - automatic_shard: true device: mi355_dpx num_devices: 1 timeout_in_minutes: 30 @@ -1201,7 +1151,6 @@ steps: - label: ":nvidia: (H200 MIG 18GB) LM Eval TurboQuant t3nc" key: lm-eval-turboquant-t3nc - automatic_shard: true timeout_in_minutes: 30 device: h200_18gb source_file_dependencies: @@ -1215,7 +1164,6 @@ steps: amd: label: ":amd: (MI355 DPX) LM Eval TurboQuant t3nc" dind: false - automatic_shard: true device: mi355_dpx num_devices: 1 timeout_in_minutes: 30 @@ -1244,7 +1192,6 @@ steps: # Octave kernels are built only for gfx942 and gfx1100, so this runs on MI300. - label: ":amd: (MI300 DPX) LM Eval Octave KV Cache" key: amd-lm-eval-octave-mi300 - automatic_shard: true dind: false device: mi300_dpx num_devices: 1 @@ -1265,7 +1212,6 @@ steps: - label: ":nvidia: (H100) [Nightly] GPQA Eval (GPT-OSS)" key: gpqa-eval-gpt-oss-2xh100 - automatic_shard: true timeout_in_minutes: 35 device: h100 optional: true @@ -1279,7 +1225,6 @@ steps: mirror: amd: label: ":amd: (MI300) GPQA Eval (GPT-OSS)" - automatic_shard: true dind: false device: mi300_2 num_devices: 2 @@ -1303,7 +1248,6 @@ steps: - label: ":nvidia: (B200) [Nightly] GPQA Eval (GPT-OSS)" key: gpqa-eval-gpt-oss-2xb200 - automatic_shard: true timeout_in_minutes: 30 device: b200-k8s optional: true @@ -1316,7 +1260,6 @@ steps: - pytest -s -v evals/gpt_oss/test_gpqa_correctness.py --config-list-file=configs/models-b200.txt mirror: amd: - automatic_shard: true label: ":amd: (MI355) GPQA Eval (GPT-OSS)" dind: false device: mi355_2 @@ -1419,7 +1362,6 @@ steps: - label: ":nvidia: (DGX Spark) [Nightly] GPQA Eval (GPT-OSS)" key: gpqa-eval-gpt-oss-spark - automatic_shard: true timeout_in_minutes: 35 device: dgx-spark optional: true @@ -1434,7 +1376,6 @@ steps: - pytest -s -v evals/gpt_oss/test_gpqa_correctness.py --config-list-file=configs/models-spark.txt mirror: amd: - automatic_shard: true label: ":amd: (MI355 DPX) GPQA Eval (GPT-OSS TP1)" # The Spark config is a backend-neutral GPT-OSS TP1 baseline. dind: false @@ -1458,7 +1399,6 @@ steps: - label: ":nvidia: (H200 MIG 35GB) LM Eval KV-Offload" key: kv-offload-small - automatic_shard: true timeout_in_minutes: 30 device: h200_35gb source_file_dependencies: @@ -1471,7 +1411,6 @@ steps: - pytest -s -v evals/gsm8k/test_gsm8k_offloading.py -k "nemotron-h-8b or gemma-4-e4b-it" mirror: amd: - automatic_shard: true label: ":amd: (MI355 DPX) LM Eval KV-Offload" dind: false device: mi355_dpx @@ -1491,7 +1430,6 @@ steps: - label: ":nvidia: (H100) LM Eval KV-Offload Medium" key: kv-offload-medium - automatic_shard: true timeout_in_minutes: 45 device: h100 num_devices: 2 @@ -1505,7 +1443,6 @@ steps: - pytest -s -v evals/gsm8k/test_gsm8k_offloading.py -k "qwen3.5-35b or deepseek-v2-lite" mirror: amd: - automatic_shard: true label: ":amd: (MI355) LM Eval KV-Offload Medium" dind: false device: mi355_2 @@ -1523,7 +1460,6 @@ steps: - label: ":nvidia: (H100) [Nightly] LM Eval KV-Offload Large" key: kv-offload-large - automatic_shard: true # Nightly only: 4-GPU H100 pods wait behind smaller jobs on per-PR runs. optional: true timeout_in_minutes: 40 @@ -1539,7 +1475,6 @@ steps: - pytest -s -v evals/gsm8k/test_gsm8k_offloading.py -k "deepseek-v4-flash" mirror: amd: - automatic_shard: true label: ":amd: (MI355) LM Eval KV-Offload Large" dind: false device: mi355_4 @@ -1559,7 +1494,6 @@ steps: - label: ":nvidia: (H200 MIG 35GB) MRCR Eval Small Models" key: mrcr-eval-small-models - automatic_shard: true device: h200_35gb timeout_in_minutes: 25 source_file_dependencies: @@ -1568,7 +1502,6 @@ steps: - pytest -s -v evals/mrcr/test_mrcr_correctness.py --config-list-file=evals/mrcr/configs/models-small.txt mirror: amd: - automatic_shard: true label: ":amd: (MI355 DPX) MRCR Eval Small Models" dind: false device: mi355_dpx @@ -1586,7 +1519,6 @@ steps: - label: ":amd: (MI355) [Nightly] LM Eval Large Models Harness" key: amd-lm-eval-large-models-harness - automatic_shard: true dind: false device: mi355_8 optional: true @@ -1612,7 +1544,6 @@ steps: - label: ":amd: (MI355) [Nightly] LM Eval Large Models ROCm Harness" key: amd-lm-eval-large-models-rocm-harness - automatic_shard: true dind: false device: mi355_8 optional: true diff --git a/.buildkite/test_areas/lora.yaml b/.buildkite/test_areas/lora.yaml index 74770b4a4d89..70ea24f82d3f 100644 --- a/.buildkite/test_areas/lora.yaml +++ b/.buildkite/test_areas/lora.yaml @@ -5,7 +5,6 @@ steps: - label: ":nvidia: (H200 MIG 35GB) LoRA" device: h200_35gb key: lora - automatic_shard: true timeout_in_minutes: 40 source_file_dependencies: - vllm/lora @@ -28,7 +27,6 @@ steps: mirror: amd: label: ":amd: (MI355 DPX) LoRA" - automatic_shard: true dind: false device: mi355_dpx working_dir: "/vllm-workspace/tests" @@ -43,7 +41,6 @@ steps: - label: ":nvidia: (L4) [Nightly] LoRA TP (Distributed)" key: lora-tp-distributed - automatic_shard: true # Nightly only: 4-GPU L4 pods wait behind smaller jobs on per-PR runs. optional: true timeout_in_minutes: 25 @@ -74,7 +71,6 @@ steps: amd: # Run on PRs even when the L4 step is optional (nightly only). optional: false - automatic_shard: true label: ":amd: (MI300) LoRA TP (Distributed)" dind: false device: mi300_4 diff --git a/.buildkite/test_areas/misc.yaml b/.buildkite/test_areas/misc.yaml index 6b66403813ad..d2b8f18a36e8 100644 --- a/.buildkite/test_areas/misc.yaml +++ b/.buildkite/test_areas/misc.yaml @@ -4,7 +4,6 @@ depends_on: steps: - label: ":nvidia: (H200 MIG 18GB) V1 Sample" key: v1-sample - automatic_shard: true timeout_in_minutes: 45 device: h200_18gb source_file_dependencies: &v1-sample-logits-deps @@ -32,7 +31,6 @@ steps: amd: label: ":amd: (MI355 DPX) V1 Sample" dind: false - automatic_shard: true device: mi355_dpx timeout_in_minutes: 70 depends_on: @@ -40,7 +38,6 @@ steps: - label: ":nvidia: (H200 MIG 18GB) V1 Logits + Oracle" key: v1-logits-oracle - automatic_shard: true timeout_in_minutes: 45 device: h200_18gb source_file_dependencies: *v1-sample-logits-deps @@ -56,7 +53,6 @@ steps: amd: label: ":amd: (MI355 DPX) V1 Logits + Oracle" dind: false - automatic_shard: true device: mi355_dpx timeout_in_minutes: 70 depends_on: @@ -65,7 +61,6 @@ steps: - label: ":nvidia: (H200 MIG 35GB) V1 Core" device: h200_35gb key: v1-core - automatic_shard: true timeout_in_minutes: 45 source_file_dependencies: &v1-core-kv-metrics-deps - vllm/config/ @@ -101,7 +96,6 @@ steps: - pytest -v -s -m 'not cpu_test' v1/core mirror: amd: - automatic_shard: true label: ":amd: (MI355 DPX) V1 Core" dind: false device: mi355_dpx @@ -112,8 +106,10 @@ steps: - label: ":nvidia: (H200 MIG 18GB) V1 Executor + Worker" device: h200_18gb key: v1-executor-worker - automatic_shard: true timeout_in_minutes: 45 + env: + # staging: pytest-shard flags in the environment keep the job whole (one shard of one). + PYTEST_ADDOPTS: "--num-shards=1 --shard-id=0" source_file_dependencies: *v1-core-kv-metrics-deps commands: - export VLLM_WORKER_MULTIPROC_METHOD=spawn @@ -125,7 +121,6 @@ steps: amd: label: ":amd: (MI355 DPX) V1 Executor + Worker" dind: false - automatic_shard: true device: mi355_dpx timeout_in_minutes: 60 depends_on: @@ -134,7 +129,6 @@ steps: - label: ":nvidia: (H200 MIG 35GB) V1 KV Offload" device: h200_35gb key: v1-kv-offload - automatic_shard: true timeout_in_minutes: 45 source_file_dependencies: *v1-core-kv-metrics-deps commands: @@ -144,7 +138,6 @@ steps: - pytest -v -s v1/simple_kv_offload mirror: amd: - automatic_shard: true label: ":amd: (MI355 DPX) V1 KV Offload" dind: false device: mi355_dpx @@ -155,7 +148,6 @@ steps: - label: ":nvidia: (H200 MIG 35GB) V1 KV Connectors" device: h200_35gb key: v1-kv-connectors - automatic_shard: true timeout_in_minutes: 45 source_file_dependencies: *v1-core-kv-metrics-deps commands: @@ -166,7 +158,6 @@ steps: mirror: amd: label: ":amd: (MI355 DPX) V1 KV Connectors" - automatic_shard: true dind: false device: mi355_dpx timeout_in_minutes: 60 @@ -178,7 +169,6 @@ steps: # MoRIIO suite on ROCm whenever the connector or its tests change. - label: ":amd: (MI355 DPX) MoRIIO Connector Unit Tests" key: amd-moriio-connector-unit-mi355 - automatic_shard: true dind: false device: mi355_dpx num_devices: 1 @@ -203,7 +193,6 @@ steps: # the connector or its tests change. - label: ":amd: (MI355 DPX) UMBP Connector Unit Tests" key: amd-umbp-connector-unit-mi355 - automatic_shard: true dind: false device: mi355_dpx num_devices: 1 @@ -230,7 +219,6 @@ steps: - label: ":amd: (MI355 DPX) UMBP Connector Integration" key: amd-umbp-connector-integration-mi355 - automatic_shard: true dind: false device: mi355_dpx num_devices: 1 @@ -254,7 +242,6 @@ steps: - label: ":amd: (MI355) UMBP Connector TP2 Integration" key: amd-umbp-connector-tp2-integration-mi355 - automatic_shard: true dind: false device: mi355_2 num_devices: 2 @@ -281,7 +268,6 @@ steps: # decode on separate GPUs over xGMI, compared exactly against a direct serve. - label: ":amd: (MI355) MoRIIO WRITE P/D Dummy-Weight E2E" key: amd-moriio-write-dummy-e2e-mi355 - automatic_shard: true dind: false device: mi355_2 num_devices: 2 @@ -308,7 +294,6 @@ steps: # TP1 <-> TP2), which needs 3 GPUs: 8 cases, ~36 min on MI355. - label: ":amd: (MI355) MoRIIO WRITE P/D Dummy-Weight Heterogeneous TP" key: amd-moriio-write-dummy-hetero-tp-mi355 - automatic_shard: true dind: false device: mi355_4 num_devices: 4 @@ -332,7 +317,6 @@ steps: - label: ":nvidia: (H200 MIG 35GB) V1 Metrics + LM Eval" device: h200_35gb key: v1-metrics-lmeval - automatic_shard: true timeout_in_minutes: 45 env: # GitHub intermittently rejects HTTP/2 upload-pack requests from the H200 @@ -350,7 +334,6 @@ steps: - pytest -v -s entrypoints/openai/correctness/test_lmeval.py::test_lm_eval_accuracy_v1_engine mirror: amd: - automatic_shard: true label: ":amd: (MI355 DPX) V1 Metrics + LM Eval" dind: false device: mi355_dpx @@ -398,7 +381,6 @@ steps: - label: ":nvidia: (H200 MIG 18GB) Extract Hidden States Integration" key: extract-hidden-states-integration - automatic_shard: true timeout_in_minutes: 20 device: h200_18gb working_dir: "/vllm-workspace" @@ -412,7 +394,6 @@ steps: - pytest -v -s tests/v1/kv_connector/extract_hidden_states_integration mirror: amd: - automatic_shard: true label: ":amd: (MI355 DPX) Extract Hidden States Integration Single GPU" dind: false device: mi355_dpx @@ -434,7 +415,6 @@ steps: - label: ":nvidia: (L4) Extract Hidden States Integration" key: extract-hidden-states-integration-2-gpus - automatic_shard: true timeout_in_minutes: 20 device: l4 num_devices: 2 @@ -449,7 +429,6 @@ steps: - pytest -v -s -m 'distributed' tests/v1/kv_connector/extract_hidden_states_integration mirror: amd: - automatic_shard: true label: ":amd: (MI355) Extract Hidden States Integration" dind: false device: mi355_2 @@ -513,7 +492,6 @@ steps: - label: ":nvidia: (H200 MIG 35GB) Examples" device: h200_35gb key: examples - automatic_shard: true timeout_in_minutes: 40 working_dir: "/vllm-workspace/tests" source_file_dependencies: @@ -527,7 +505,6 @@ steps: mirror: amd: label: ":amd: (MI355 DPX) Examples" - automatic_shard: true dind: false device: mi355_dpx timeout_in_minutes: 75 @@ -542,7 +519,6 @@ steps: - label: ":nvidia: (L4) Metrics, Tracing" key: metrics-tracing-2-gpus - automatic_shard: true timeout_in_minutes: 25 device: l4 num_devices: 2 @@ -567,7 +543,6 @@ steps: - pytest -v -s tracing mirror: amd: - automatic_shard: true label: ":amd: (MI300) Metrics, Tracing" dind: false device: mi300_2 @@ -624,7 +599,6 @@ steps: - label: ":nvidia: (H200 MIG 35GB) Async Engine, Inputs, Utils, Worker" device: h200_35gb key: async-engine-inputs-utils-worker - automatic_shard: true timeout_in_minutes: 25 source_file_dependencies: - vllm/assets/ @@ -651,8 +625,8 @@ steps: mirror: amd: label: ":amd: (MI355 DPX) Async Engine, Inputs, Utils, Worker" + automatic_shard: false # staging: a mirror that opts out alone dind: false - automatic_shard: true device: mi355_dpx timeout_in_minutes: 50 depends_on: @@ -915,7 +889,6 @@ steps: # Key kept from the former A100 step's mirror for job history and retries. - label: ":amd: (MI250) Batch Invariance" key: amd-batch-invariance-a100 - automatic_shard: true dind: false device: mi250_1 timeout_in_minutes: 60 @@ -940,7 +913,6 @@ steps: - label: ":nvidia: (H100) [Nightly] Batch Invariance" key: batch-invariance-h100 - automatic_shard: true timeout_in_minutes: 60 device: h100 optional: true @@ -964,7 +936,6 @@ steps: optional: false label: ":amd: (MI300) Batch Invariance" dind: false - automatic_shard: true device: mi300_1 timeout_in_minutes: 60 depends_on: @@ -990,7 +961,6 @@ steps: - label: ":nvidia: (B200) Batch Invariance" key: batch-invariance-b200 - automatic_shard: true timeout_in_minutes: 45 device: b200-k8s source_file_dependencies: @@ -1016,7 +986,6 @@ steps: amd: label: ":amd: (MI355) Batch Invariance" dind: false - automatic_shard: true device: mi355_1 timeout_in_minutes: 60 depends_on: @@ -1046,7 +1015,6 @@ steps: - label: ":nvidia: (H100) [Nightly] Batch Invariance (Distributed)" key: batch-invariance-distributed-4xh100 - automatic_shard: true timeout_in_minutes: 45 device: h100 optional: true @@ -1067,7 +1035,6 @@ steps: - VLLM_TEST_MODEL=ibm-granite/granite-3.0-1b-a400m-base VLLM_TEST_SERVER_ARGS="--data-parallel-size 4 --enable-expert-parallel --max-model-len 4096" pytest -v -s v1/determinism/test_online_batch_invariance.py mirror: amd: - automatic_shard: true label: ":amd: (MI300) Batch Invariance (Distributed)" dind: false device: mi300_4 @@ -1093,7 +1060,6 @@ steps: - label: ":nvidia: (H200 MIG 35GB) [Nightly] Acceptance Length (Large Models)" device: h200_35gb key: acceptance-length-test-large-models - automatic_shard: true timeout_in_minutes: 20 gpu: h100 optional: true @@ -1107,7 +1073,6 @@ steps: - pytest -v -s v1/spec_decode/test_acceptance_length.py -m slow_test mirror: amd: - automatic_shard: true label: ":amd: (MI355 DPX) Acceptance Length (Large Models)" dind: false device: mi355_dpx diff --git a/.buildkite/test_areas/model_executor.yaml b/.buildkite/test_areas/model_executor.yaml index cfa925b2d4c9..67f42abac844 100644 --- a/.buildkite/test_areas/model_executor.yaml +++ b/.buildkite/test_areas/model_executor.yaml @@ -60,7 +60,6 @@ steps: - label: ':amd: (MI300) Model Executor' device: mi300_1 key: amd-model-executor - automatic_shard: true timeout_in_minutes: 75 source_file_dependencies: - vllm/engine/arg_utils.py diff --git a/.buildkite/test_areas/models_basic.yaml b/.buildkite/test_areas/models_basic.yaml index 04bc43a2a13b..b013897023a0 100644 --- a/.buildkite/test_areas/models_basic.yaml +++ b/.buildkite/test_areas/models_basic.yaml @@ -4,7 +4,6 @@ depends_on: steps: - label: ":nvidia: (H200 MIG 18GB) Basic Models (Initialization)" key: basic-models-tests-initialization - automatic_shard: true timeout_in_minutes: 25 device: h200_18gb source_file_dependencies: @@ -24,7 +23,6 @@ steps: - pytest -v -s v1/spec_decode/test_dflash2.py -k "beam_conv_residual_norm or grouped_conv_triton_matches_reference or beam_fused_candidate_edges" mirror: amd: - automatic_shard: true label: ":amd: (MI355 DPX) Basic Models (Initialization)" dind: false device: mi355_dpx @@ -35,7 +33,6 @@ steps: - label: ":nvidia: (H200 MIG 35GB) Basic Models (Extra Initialization)" device: h200_35gb key: basic-models-tests-extra-initialization - automatic_shard: true timeout_in_minutes: 30 source_file_dependencies: - vllm/model_executor/models/ @@ -49,7 +46,6 @@ steps: mirror: amd: label: ":amd: (MI355 DPX) Basic Models (Extra Initialization)" - automatic_shard: true dind: false device: mi355_dpx timeout_in_minutes: 60 @@ -67,7 +63,6 @@ steps: - label: ":nvidia: (H200 MIG 35GB) Basic Models (Other)" device: h200_35gb key: basic-models-tests-other - automatic_shard: true timeout_in_minutes: 38 source_file_dependencies: - vllm/ @@ -115,7 +110,6 @@ steps: amd: label: ":amd: (MI355) Basic Models (Other)" dind: false - automatic_shard: true device: mi355_1 timeout_in_minutes: 65 depends_on: @@ -181,7 +175,6 @@ steps: - label: ":nvidia: (B200) Inkling" key: inkling-unit-tests-b200 - automatic_shard: true timeout_in_minutes: 40 device: b200-k8s source_file_dependencies: @@ -194,7 +187,6 @@ steps: - pytest -v -s models/inkling mirror: amd: - automatic_shard: true label: ":amd: (MI355) Inkling" dind: false device: mi355_1 @@ -226,7 +218,6 @@ steps: - label: ":nvidia: (B200) Kimi K3" key: kimi-k3-unit-tests-b200 - automatic_shard: true timeout_in_minutes: 40 device: b200-k8s source_file_dependencies: @@ -248,7 +239,6 @@ steps: kernels/attention/test_kimi_k3_mla_fused_epilogue.py mirror: amd: - automatic_shard: true label: ":amd: (MI355) Kimi K3" dind: false device: mi355_1 @@ -280,7 +270,6 @@ steps: key: kimi-k3-prefix-cache-4xb200 # Nightly only: 4-GPU B200 pods wait hours behind smaller jobs on per-PR runs. optional: true - automatic_shard: true timeout_in_minutes: 75 device: b200-k8s num_devices: 4 @@ -306,7 +295,6 @@ steps: mirror: amd: label: ":amd: (MI355) Kimi K3 Prefix Cache" - automatic_shard: true dind: false device: mi355_4 timeout_in_minutes: 60 @@ -322,7 +310,6 @@ steps: - label: ":nvidia: (H200) Step5" key: step5-unit-tests - automatic_shard: true timeout_in_minutes: 20 device: h200_35gb source_file_dependencies: @@ -333,7 +320,6 @@ steps: - pytest -v -s models/step5 mirror: amd: - automatic_shard: true label: ":amd: (MI355 DPX) Step5" dind: false device: mi355_dpx @@ -363,7 +349,6 @@ steps: - label: ":nvidia: (H200) GLM5Next" key: glm5next-unit-tests - automatic_shard: true timeout_in_minutes: 20 device: h200_35gb source_file_dependencies: @@ -374,7 +359,6 @@ steps: - pytest -v -s models/glm5next mirror: amd: - automatic_shard: true label: ":amd: (MI355 DPX) GLM5Next" dind: false device: mi355_dpx @@ -387,7 +371,6 @@ steps: - label: ":nvidia: (H200) Qwen4 Exp" key: qwen4-exp-unit-tests - automatic_shard: true timeout_in_minutes: 40 device: h200_35gb source_file_dependencies: @@ -405,7 +388,6 @@ steps: - pytest -v -s models/qwen4_exp/test_ple.py mirror: amd: - automatic_shard: true label: ":amd: (MI355 DPX) Qwen4 Exp" dind: false device: mi355_dpx @@ -422,7 +404,6 @@ steps: - label: ":nvidia: (B200) Qwen4 Exp Kernels" key: qwen4-exp-kernels-b200 - automatic_shard: true timeout_in_minutes: 20 device: b200-k8s source_file_dependencies: @@ -443,7 +424,6 @@ steps: mirror: amd: label: ":amd: (MI355) Qwen4 Exp HC Kernels" - automatic_shard: true dind: false device: mi355_1 depends_on: @@ -458,7 +438,6 @@ steps: - label: ":amd: (MI300) Qwen4 Exp" key: qwen4-exp-unit-tests-mi300 - automatic_shard: true timeout_in_minutes: 40 dind: false device: mi300_1 @@ -495,7 +474,6 @@ steps: - label: ":nvidia: (B200) Qwen3.5" key: qwen3-5-unit-tests-b200 - automatic_shard: true timeout_in_minutes: 20 device: b200-k8s # test_fused_allreduce_norm_quant runs TP2. @@ -567,7 +545,6 @@ steps: - label: ":nvidia: (4xB200) DeepSeek-V4.1 NVIDIA Kernels" key: dsv41-nvidia-kernels-4xb200 - automatic_shard: true # Nightly only: 4-GPU B200 pods wait hours behind smaller jobs on per-PR runs. optional: true timeout_in_minutes: 50 @@ -606,7 +583,6 @@ steps: - label: ":nvidia: (4xB200) GLM-5.3-Flash Specialized Decode" key: glm5next-nvidia-specialized-decode-4xb200 - automatic_shard: true # Nightly only: 4-GPU B200 pods wait hours behind smaller jobs on per-PR runs. optional: true timeout_in_minutes: 40 diff --git a/.buildkite/test_areas/models_distributed.yaml b/.buildkite/test_areas/models_distributed.yaml index 4b49f802988e..bbd1bd546d12 100644 --- a/.buildkite/test_areas/models_distributed.yaml +++ b/.buildkite/test_areas/models_distributed.yaml @@ -4,7 +4,6 @@ depends_on: steps: - label: ":nvidia: (L4) [Nightly] Qwen4 Exp SP" key: qwen4-exp-sp-tests-4-gpus - automatic_shard: true # Nightly only: 4-GPU L4 pods wait behind smaller jobs on per-PR runs. optional: true timeout_in_minutes: 30 @@ -23,7 +22,6 @@ steps: - pytest -v -s models/qwen4_exp/test_tpsp.py mirror: amd: - automatic_shard: true # Run on PRs even when the L4 step is optional (nightly only). optional: false label: ":amd: (MI355) Qwen4 Exp SP" @@ -43,7 +41,6 @@ steps: - label: ":nvidia: (L4) Distributed Models" key: distributed-model-tests-2-gpus - automatic_shard: true timeout_in_minutes: 35 working_dir: "/vllm-workspace/tests" device: l4 @@ -65,7 +62,6 @@ steps: - VLLM_TRITON_FORCE_FIRST_CONFIG=1 pytest -v -s models/test_pp_equivalence.py -m 'distributed(num_gpus=2)' mirror: amd: - automatic_shard: true label: ":amd: (MI300) Distributed Models" dind: false device: mi300_2 @@ -105,7 +101,6 @@ steps: # cases that run on PRs; 4-GPU L4 pods wait behind smaller jobs on PR runs. # automatic_shard needs timings from a passing run; until then this is one job. optional: true - automatic_shard: true timeout_in_minutes: 240 working_dir: "/vllm-workspace/tests" device: l4 diff --git a/.buildkite/test_areas/models_language.yaml b/.buildkite/test_areas/models_language.yaml index 2ceb354776dd..8f7f8ec2ef53 100644 --- a/.buildkite/test_areas/models_language.yaml +++ b/.buildkite/test_areas/models_language.yaml @@ -4,7 +4,7 @@ depends_on: steps: - label: ":nvidia: (H200 MIG 18GB) Language Models (Standard)" key: language-models-tests-standard - automatic_shard: true + automatic_shard: false # staging: a step that opts out while its mirror opts in timeout_in_minutes: 30 device: h200_18gb source_file_dependencies: @@ -19,7 +19,7 @@ steps: mirror: amd: label: ":amd: (MI355 DPX) Language Models (Standard)" - automatic_shard: true + automatic_shard: true # staging: a mirror that opts in alone dind: false device: mi355_dpx timeout_in_minutes: 60 @@ -29,7 +29,6 @@ steps: - label: ":nvidia: (H200 MIG 35GB) Language Models (Extra Standard)" device: h200_35gb key: language-models-tests-extra-standard - automatic_shard: true timeout_in_minutes: 40 source_file_dependencies: - vllm/model_executor/models/ @@ -44,7 +43,6 @@ steps: mirror: amd: label: ":amd: (MI355 DPX) Language Models (Extra Standard)" - automatic_shard: true dind: false device: mi355_dpx timeout_in_minutes: 40 @@ -64,7 +62,6 @@ steps: - label: ":nvidia: (H200 MIG 35GB) Language Models (Hybrid)" device: h200_35gb key: language-models-tests-hybrid - automatic_shard: true timeout_in_minutes: 65 source_file_dependencies: - vllm/ @@ -80,7 +77,6 @@ steps: - pytest -v -s models/language/generation -m hybrid_model -k 'not granite-4.0-tiny-preview' mirror: amd: - automatic_shard: true label: ":amd: (MI355 DPX) Language Models (Hybrid)" dind: false device: mi355_dpx @@ -95,7 +91,6 @@ steps: # correctness test on L4 until its H200 output matches the Transformers reference. - label: ":nvidia: (L4) Granite Language Model Compatibility" key: language-models-tests-granite-l4-compatibility - automatic_shard: true device: l4 num_devices: 1 timeout_in_minutes: 65 @@ -112,7 +107,6 @@ steps: - pytest -v -s models/language/generation -m hybrid_model -k 'granite-4.0-tiny-preview' mirror: amd: - automatic_shard: true label: ":amd: (MI355 DPX) Granite Language Model Compatibility" dind: false device: mi355_dpx @@ -131,7 +125,6 @@ steps: - label: ":nvidia: (H200 MIG 35GB) [Nightly] Language Models (Extended Generation)" device: h200_35gb key: language-models-test-extended-generation - automatic_shard: true timeout_in_minutes: 80 optional: true source_file_dependencies: @@ -153,7 +146,6 @@ steps: --ignore=models/language/generation/test_gdn_sleep_wake.py mirror: amd: - automatic_shard: true label: ":amd: (MI355 DPX) Language Models (Extended Generation)" dind: false device: mi355_dpx @@ -170,7 +162,6 @@ steps: - label: ":nvidia: (H200 MIG 18GB) [Nightly] Language Models (PPL)" key: language-models-test-ppl - automatic_shard: true timeout_in_minutes: 30 device: h200_18gb optional: true @@ -184,7 +175,6 @@ steps: amd: label: ":amd: (MI250) Language Models (PPL)" dind: false - automatic_shard: true device: mi250_1 timeout_in_minutes: 65 depends_on: @@ -211,7 +201,6 @@ steps: - label: ":nvidia: (H200 MIG 35GB) [Nightly] Language Models (Extended Pooling)" device: h200_35gb key: language-models-test-extended-pooling - automatic_shard: true timeout_in_minutes: 120 optional: true source_file_dependencies: @@ -225,7 +214,6 @@ steps: - pytest -v -s models/language/pooling -m 'not core_model' --timeout=900 --timeout-method=thread mirror: amd: - automatic_shard: true commands: - "pytest -v -s models/language/pooling -m 'not core_model'" label: ":amd: (MI355 DPX) Language Models (Extended Pooling)" @@ -237,7 +225,6 @@ steps: - label: ":nvidia: (H200 MIG 35GB) [Nightly] Language Models (MTEB)" key: language-models-test-mteb - automatic_shard: true timeout_in_minutes: 68 device: h200_35gb optional: true @@ -250,7 +237,6 @@ steps: mirror: amd: label: ":amd: (MI250) Language Models (MTEB)" - automatic_shard: true dind: false device: mi250_1 timeout_in_minutes: 60 diff --git a/.buildkite/test_areas/models_multimodal.yaml b/.buildkite/test_areas/models_multimodal.yaml index f8c3577a0226..8116363e1319 100644 --- a/.buildkite/test_areas/models_multimodal.yaml +++ b/.buildkite/test_areas/models_multimodal.yaml @@ -4,7 +4,6 @@ depends_on: steps: - label: ":nvidia: (H200 MIG 18GB) Multimodal Models (Standard) 1: qwen2" key: multi-modal-models-standard-1-qwen2 - automatic_shard: true timeout_in_minutes: 68 device: h200_18gb source_file_dependencies: @@ -18,7 +17,6 @@ steps: amd: label: ":amd: (MI355 DPX) Multimodal Models (Standard) 1: qwen2" dind: false - automatic_shard: true device: mi355_dpx timeout_in_minutes: 65 depends_on: @@ -26,7 +24,6 @@ steps: - label: ":nvidia: (H200 MIG 18GB) Multimodal Models (Standard) 2: qwen3 + gemma" key: multi-modal-models-standard-2-qwen3-gemma - automatic_shard: true timeout_in_minutes: 75 device: h200_18gb source_file_dependencies: @@ -40,7 +37,6 @@ steps: mirror: amd: label: ":amd: (MI355 DPX) Multimodal Models (Standard) 2: qwen3 + gemma" - automatic_shard: true dind: false device: mi355_dpx timeout_in_minutes: 55 @@ -50,7 +46,6 @@ steps: - label: ":nvidia: (H200 MIG 35GB) Multimodal Models (Standard) 3: llava + qwen2_vl" device: h200_35gb key: multi-modal-models-standard-3-llava-qwen2-vl - automatic_shard: true timeout_in_minutes: 40 source_file_dependencies: - vllm/ @@ -61,7 +56,6 @@ steps: - pytest -v -s models/multimodal/generation/test_qwen2_vl.py -m core_model mirror: amd: - automatic_shard: true label: ":amd: (MI250) Multimodal Models (Standard) 3: llava + qwen2_vl" dind: false device: mi250_1 @@ -72,7 +66,6 @@ steps: - label: ":nvidia: (H200 MIG 35GB) Unlimited-OCR-2" device: h200_35gb key: unlimited-ocr2-generation - automatic_shard: true timeout_in_minutes: 30 source_file_dependencies: - vllm/ @@ -84,7 +77,6 @@ steps: - label: ":nvidia: (H200 MIG 35GB) Multimodal Models (Standard) 4: other + whisper" device: h200_35gb key: multi-modal-models-standard-4-other-whisper - automatic_shard: true timeout_in_minutes: 75 source_file_dependencies: - vllm/ @@ -111,7 +103,6 @@ steps: mirror: amd: label: ":amd: (MI355 DPX) Multimodal Models (Standard) 4: other + whisper" - automatic_shard: true dind: false device: mi355_dpx timeout_in_minutes: 70 @@ -139,7 +130,6 @@ steps: - label: ":nvidia: (H200 MIG 18GB) Multimodal Processor" key: multi-modal-processor - automatic_shard: true timeout_in_minutes: 98 device: h200_18gb source_file_dependencies: @@ -152,7 +142,6 @@ steps: mirror: amd: label: ":amd: (MI355 DPX) Multimodal Processor" - automatic_shard: true dind: false device: mi355_dpx timeout_in_minutes: 115 @@ -165,7 +154,6 @@ steps: - label: ":nvidia: (H200 MIG 35GB) Multimodal Accuracy Eval (Small Models)" device: h200_35gb key: multi-modal-accuracy-eval-small-models - automatic_shard: true timeout_in_minutes: 30 working_dir: "/vllm-workspace/.buildkite/lm-eval-harness" source_file_dependencies: @@ -176,7 +164,6 @@ steps: - pytest -s -v test_lm_eval_correctness.py --config-list-file=configs/models-mm-small.txt --tp-size=1 mirror: amd: - automatic_shard: true label: ":amd: (MI355 DPX) Multimodal Accuracy Eval (Small Models)" dind: false device: mi355_dpx @@ -193,7 +180,6 @@ steps: - label: ":nvidia: (H200 MIG 35GB) [Nightly] Multimodal Models (Extended Generation 1)" device: h200_35gb key: multi-modal-models-extended-generation-1 - automatic_shard: true optional: true env: # Engines in a shard can start after other tests touched CUDA in the @@ -213,7 +199,6 @@ steps: - pytest -v -s models/multimodal/test_mapping.py mirror: amd: - automatic_shard: true label: ":amd: (MI355 DPX) Multimodal Models (Extended Generation 1)" dind: false device: mi355_dpx @@ -224,7 +209,6 @@ steps: - label: ":nvidia: (H200 MIG 35GB) [Nightly] Multimodal Models (PPL)" device: h200_35gb key: multi-modal-models-extended-ppl - automatic_shard: true optional: true source_file_dependencies: - vllm/ @@ -234,7 +218,6 @@ steps: - pytest -v -s models/multimodal/generation_ppl_test/ mirror: amd: - automatic_shard: true label: ":amd: (MI355 DPX) Multimodal Models (PPL)" dind: false device: mi355_dpx @@ -245,7 +228,6 @@ steps: - label: ":nvidia: (H200 MIG 35GB) [Nightly] Multimodal Models (Extended Generation 2)" device: h200_35gb key: multi-modal-models-extended-generation-2 - automatic_shard: true optional: true source_file_dependencies: - vllm/ @@ -256,7 +238,6 @@ steps: mirror: amd: label: ":amd: (MI355 DPX) Multimodal Models (Extended Generation 2)" - automatic_shard: true dind: false device: mi355_dpx timeout_in_minutes: 100 @@ -266,7 +247,6 @@ steps: - label: ":nvidia: (H200 MIG 35GB) [Nightly] Multimodal Models (Extended Generation 3)" device: h200_35gb key: multi-modal-models-extended-generation-3 - automatic_shard: true optional: true source_file_dependencies: - vllm/ @@ -278,7 +258,6 @@ steps: amd: label: ":amd: (MI355 DPX) Multimodal Models (Extended Generation 3)" dind: false - automatic_shard: true device: mi355_dpx timeout_in_minutes: 90 depends_on: @@ -286,7 +265,6 @@ steps: - label: ":nvidia: (H200 MIG 35GB) [Nightly] Multimodal Models (Extended Pooling)" key: multi-modal-models-extended-pooling - automatic_shard: true optional: true device: h200_35gb source_file_dependencies: @@ -299,7 +277,6 @@ steps: amd: label: ":amd: (MI355 DPX) Multimodal Models (Extended Pooling)" dind: false - automatic_shard: true device: mi355_dpx timeout_in_minutes: 75 depends_on: diff --git a/.buildkite/test_areas/monokernels.yaml b/.buildkite/test_areas/monokernels.yaml index ca7cd647d7e3..a45700ab732a 100644 --- a/.buildkite/test_areas/monokernels.yaml +++ b/.buildkite/test_areas/monokernels.yaml @@ -21,7 +21,6 @@ depends_on: steps: - label: ':amd: (MI355) DeepSeek-V4.1 Mono Decode Numerics' key: dsv41-mono-decode-numerics-amd - automatic_shard: true dind: false device: mi355_4 num_devices: 4 @@ -41,7 +40,6 @@ steps: - label: ':amd: (MI355) DeepSeek-V4.1 Mono Decode Numerics TP8' key: dsv41-mono-decode-numerics-tp8-amd - automatic_shard: true dind: false device: mi355_8 num_devices: 8 @@ -65,7 +63,6 @@ steps: - label: ':amd: (MI300) DeepSeek-V4.1 gfx942 Mono Decode Layers' key: dsv41-mono-decode-layers-gfx942-amd - automatic_shard: true dind: false device: mi300_4 num_devices: 4 @@ -85,7 +82,6 @@ steps: - label: ":amd: (MI300) DeepSeek-V4.1 gfx942 Mono Decode Kernels" key: dsv41-mono-decode-gfx942-amd - automatic_shard: true dind: false device: mi300_1 num_devices: 1 @@ -118,7 +114,6 @@ steps: - label: ':amd: (MI355) DeepSeek-V4-Pro Mono Decode Numerics' key: dsv4-mono-decode-numerics-amd - automatic_shard: true dind: false device: mi355_8 num_devices: 8 @@ -137,7 +132,6 @@ steps: - label: ":amd: (MI355) MiniMax M3 MonoKernel" key: minimax-m3-monokernel-mi355 - automatic_shard: true dind: false timeout_in_minutes: 45 device: mi355_4 @@ -162,7 +156,6 @@ steps: - label: ':amd: (MI355) Kimi-K3 Specialized Decode' key: kimi-k3-specialized-decode-amd - automatic_shard: true dind: false device: mi355_8 num_devices: 8 diff --git a/.buildkite/test_areas/plugins.yaml b/.buildkite/test_areas/plugins.yaml index e06cf687dfcc..1fad75417534 100644 --- a/.buildkite/test_areas/plugins.yaml +++ b/.buildkite/test_areas/plugins.yaml @@ -4,7 +4,6 @@ depends_on: steps: - label: ":nvidia: (L4) Plugin Integration" key: plugin-tests-2-gpus - automatic_shard: true timeout_in_minutes: 35 working_dir: "/vllm-workspace/tests" device: l4 @@ -29,7 +28,6 @@ steps: mirror: amd: label: ":amd: (MI250) Plugin Integration" - automatic_shard: true dind: false device: mi250_2 timeout_in_minutes: 50 diff --git a/.buildkite/test_areas/pytorch.yaml b/.buildkite/test_areas/pytorch.yaml index 7fdeb1f662ee..cf6b657c8a5d 100644 --- a/.buildkite/test_areas/pytorch.yaml +++ b/.buildkite/test_areas/pytorch.yaml @@ -5,7 +5,6 @@ steps: - label: ":nvidia: (H200 MIG 35GB) PyTorch Compilation" device: h200_35gb key: pytorch-compilation-unit-tests - automatic_shard: true timeout_in_minutes: 30 source_file_dependencies: &pytorch-compilation-deps - vllm/__init__.py @@ -39,7 +38,6 @@ steps: - pytest -s -v compile/ --ignore-glob='compile/*/*' mirror: amd: - automatic_shard: true label: ":amd: (MI300) PyTorch Compilation" dind: false device: mi300_1 @@ -53,14 +51,12 @@ steps: - label: ":nvidia: (H200 MIG 35GB) PyTorch Compilation / Dynamic Shapes" device: h200_35gb key: pytorch-compilation-dynamic-shapes - automatic_shard: true timeout_in_minutes: 30 source_file_dependencies: *pytorch-compilation-deps commands: - pytest -s -v compile/dynamic_shapes/ mirror: amd: - automatic_shard: true label: ":amd: (MI300) PyTorch Compilation / Dynamic Shapes" dind: false device: mi300_1 @@ -73,7 +69,6 @@ steps: - label: ":nvidia: (H200 MIG 18GB) PyTorch Compilation H100 Cases" key: pytorch-compilation-unit-tests-h100 - automatic_shard: true timeout_in_minutes: 30 device: h200_18gb source_file_dependencies: @@ -107,7 +102,6 @@ steps: - "find compile/h100/ -name 'test_*.py' -print0 | xargs -0 -n1 -I{} pytest -s -v '{}'" mirror: amd: - automatic_shard: true label: ":amd: (MI355 DPX) PyTorch Compilation H100 Cases" dind: false device: mi355_dpx @@ -123,7 +117,6 @@ steps: - label: ":nvidia: (L4) PyTorch Compilation Passes" key: pytorch-compilation-passes-unit-tests - automatic_shard: true device: l4 num_devices: 1 timeout_in_minutes: 45 @@ -163,7 +156,6 @@ steps: amd: label: ':amd: (MI300) PyTorch Compilation Passes' dind: false - automatic_shard: true device: mi300_1 timeout_in_minutes: 80 working_dir: /vllm-workspace/tests @@ -182,7 +174,6 @@ steps: - label: ":nvidia: (H200 MIG 35GB) PyTorch Fullgraph" device: h200_35gb key: pytorch-fullgraph-test - automatic_shard: true timeout_in_minutes: 90 source_file_dependencies: - vllm/__init__.py @@ -217,7 +208,6 @@ steps: - pytest -v -s compile/fullgraph --ignore=compile/fullgraph/test_full_cudagraph.py mirror: amd: - automatic_shard: true label: ":amd: (MI355 DPX) PyTorch Fullgraph" dind: false device: mi355_dpx @@ -234,7 +224,6 @@ steps: # those SM90 cases skip while the architecture-compatible cases still run. - label: ":nvidia: (L4) PyTorch Fullgraph CUDAGraph Compatibility" key: pytorch-fullgraph-cudagraph-l4-compatibility - automatic_shard: true device: l4 num_devices: 1 timeout_in_minutes: 60 @@ -269,7 +258,6 @@ steps: - pytest -s -v compile/fullgraph/test_full_cudagraph.py mirror: amd: - automatic_shard: true label: ':amd: (MI250) PyTorch Fullgraph CUDAGraph Compatibility' dind: false device: mi250_1 @@ -290,7 +278,6 @@ steps: - label: ":nvidia: (H200 MIG 18GB) PyTorch Nightly Dependency Override Check" key: pytorch-nightly-dependency-override-check - automatic_shard: true # 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 @@ -302,7 +289,6 @@ steps: - bash standalone_tests/pytorch_nightly_dependency.sh mirror: amd: - automatic_shard: true label: ":amd: (MI300 DPX) PyTorch Nightly Dependency Override Check" dind: false device: mi300_dpx diff --git a/.buildkite/test_areas/quantization.yaml b/.buildkite/test_areas/quantization.yaml index 0043d1d6f1cf..d83fc9ea49e2 100644 --- a/.buildkite/test_areas/quantization.yaml +++ b/.buildkite/test_areas/quantization.yaml @@ -56,7 +56,6 @@ steps: - label: ":nvidia: (H200 MIG 35GB) Quantized Fusions" device: h200_35gb key: quantized-fusions - automatic_shard: true timeout_in_minutes: 20 source_file_dependencies: - tests/fusion @@ -68,7 +67,6 @@ steps: - pytest -v -s fusion/ mirror: amd: - automatic_shard: true label: ":amd: (MI300 DPX) Quantized Fusions" dind: false device: mi300_dpx @@ -86,7 +84,6 @@ steps: - label: ":nvidia: (B200) Quantized MoE" key: quantized-moe-test-b200 - automatic_shard: true timeout_in_minutes: 120 working_dir: "/vllm-workspace/" device: b200-k8s @@ -121,7 +118,6 @@ steps: tests/benchmarks/test_forward_cli.py::test_bench_forward_muse_synthetic_routing mirror: amd: - automatic_shard: true label: ":amd: (MI355) Quantized MoE" dind: false device: mi355_1 @@ -152,7 +148,6 @@ steps: - label: ":nvidia: (H200 MIG 35GB) Quantized Models" device: h200_35gb key: quantized-models-test - automatic_shard: true timeout_in_minutes: 65 source_file_dependencies: - vllm/model_executor/layers/quantization @@ -163,7 +158,6 @@ steps: amd: label: ":amd: (MI355 DPX) Quantized Models" dind: false - automatic_shard: true device: mi355_dpx timeout_in_minutes: 80 working_dir: "/vllm-workspace/tests" diff --git a/.buildkite/test_areas/rl.yaml b/.buildkite/test_areas/rl.yaml index 4025f695ab1a..e9f7ac6465b3 100644 --- a/.buildkite/test_areas/rl.yaml +++ b/.buildkite/test_areas/rl.yaml @@ -5,7 +5,6 @@ steps: - label: ":nvidia: (H200 MIG 35GB) RL Entrypoints" device: h200_35gb key: rl-entrypoints-tests - automatic_shard: true timeout_in_minutes: 40 working_dir: "/vllm-workspace/tests" source_file_dependencies: @@ -20,7 +19,6 @@ steps: --ignore=entrypoints/rl/weight_transfer mirror: amd: - automatic_shard: true label: ":amd: (MI355 DPX) RL Entrypoints" dind: false device: mi355_dpx diff --git a/.buildkite/test_areas/rust_frontend.yaml b/.buildkite/test_areas/rust_frontend.yaml index 934af7877211..75534b59cd3e 100644 --- a/.buildkite/test_areas/rust_frontend.yaml +++ b/.buildkite/test_areas/rust_frontend.yaml @@ -4,7 +4,6 @@ depends_on: steps: - label: ":nvidia: (H200 MIG 18GB) Rust Frontend OpenAI Coverage" key: rust-frontend-openai-coverage - automatic_shard: true timeout_in_minutes: 50 device: h200_18gb working_dir: "/vllm-workspace/tests" @@ -65,7 +64,6 @@ steps: mirror: amd: label: ":amd: (MI355 DPX) Rust Frontend OpenAI Coverage" - automatic_shard: true dind: false device: mi355_dpx timeout_in_minutes: 55 @@ -179,7 +177,6 @@ steps: - label: ":nvidia: (H200 MIG 18GB) Rust Frontend Core Correctness" key: rust-frontend-core-correctness - automatic_shard: true timeout_in_minutes: 20 device: h200_18gb working_dir: "/vllm-workspace/tests" @@ -196,7 +193,6 @@ steps: - pytest -v -s v1/core/prefix_cache/test_hybrid_prefix_cache_hit_rate.py mirror: amd: - automatic_shard: true label: ":amd: (MI355 DPX) Rust Frontend Core Correctness" dind: false device: mi355_dpx @@ -243,7 +239,6 @@ steps: - label: ":nvidia: (H200 MIG 35GB) Rust Frontend Tool Use" key: rust-frontend-tool-use - automatic_shard: true device: h200_35gb timeout_in_minutes: 35 working_dir: "/vllm-workspace/tests" @@ -270,7 +265,6 @@ steps: - pytest -v -s entrypoints/tool_parsers/test_hermes_tool_parser.py mirror: amd: - automatic_shard: true label: ":amd: (MI355 DPX) Rust Frontend Tool Use" dind: false device: mi355_dpx @@ -290,7 +284,6 @@ steps: - label: ":nvidia: (H200 MIG 35GB) Rust Frontend Responses API" key: rust-frontend-responses-api - automatic_shard: true device: h200_35gb timeout_in_minutes: 35 working_dir: "/vllm-workspace/tests" @@ -314,7 +307,6 @@ steps: - pytest -v -s entrypoints/openai/responses/test_simple.py -k "not enable_response_messages and not streaming_logprobs" mirror: amd: - automatic_shard: true label: ":amd: (MI355 DPX) Rust Frontend Responses API" dind: false device: mi355_dpx @@ -330,7 +322,6 @@ steps: - label: ":nvidia: (L4) [Nightly] Rust Frontend Distributed" key: rust-frontend-distributed - automatic_shard: true # Nightly only: 4-GPU L4 pods wait behind smaller jobs on per-PR runs. optional: true timeout_in_minutes: 25 @@ -365,7 +356,6 @@ steps: # Run on PRs even when the L4 step is optional (nightly only). optional: false label: ":amd: (MI355) Rust Frontend Distributed" - automatic_shard: true dind: false device: mi355_4 timeout_in_minutes: 45 diff --git a/.buildkite/test_areas/samplers.yaml b/.buildkite/test_areas/samplers.yaml index 68055d6a0fd7..9d66e785ea8a 100644 --- a/.buildkite/test_areas/samplers.yaml +++ b/.buildkite/test_areas/samplers.yaml @@ -5,7 +5,6 @@ steps: - label: ":nvidia: (H200 MIG 18GB) Samplers" device: h200_18gb key: samplers-test - automatic_shard: true timeout_in_minutes: 60 source_file_dependencies: - vllm/model_executor/layers @@ -20,7 +19,6 @@ steps: - VLLM_USE_FLASHINFER_SAMPLER=1 pytest -v -s samplers -k "not test_beam_search_passes_multimodal_data" mirror: amd: - automatic_shard: true label: ":amd: (MI250) Samplers" dind: false device: mi250_1 @@ -43,7 +41,6 @@ steps: - label: ":nvidia: (H200 MIG 35GB) Samplers Multimodal Beam Search" device: h200_35gb key: samplers-multimodal-beam-search - automatic_shard: true timeout_in_minutes: 30 source_file_dependencies: - vllm/model_executor/layers @@ -55,7 +52,6 @@ steps: - VLLM_USE_FLASHINFER_SAMPLER=1 pytest -v -s samplers/test_beam_search.py::test_beam_search_passes_multimodal_data mirror: amd: - automatic_shard: true label: ":amd: (MI355 DPX) Samplers Multimodal Beam Search" dind: false device: mi355_dpx diff --git a/.buildkite/test_areas/spec_decode.yaml b/.buildkite/test_areas/spec_decode.yaml index 43b174a8604a..9d544b3d4e4b 100644 --- a/.buildkite/test_areas/spec_decode.yaml +++ b/.buildkite/test_areas/spec_decode.yaml @@ -5,7 +5,6 @@ steps: - label: ":nvidia: (H200 MIG 35GB) V1 Spec Decode" device: h200_35gb key: v1-spec-decode - automatic_shard: true timeout_in_minutes: 40 source_file_dependencies: - vllm/config/ @@ -28,7 +27,6 @@ steps: amd: label: ":amd: (MI300) V1 Spec Decode" dind: false - automatic_shard: true device: mi300_1 timeout_in_minutes: 50 depends_on: @@ -36,7 +34,6 @@ steps: - label: ":nvidia: (H200 MIG 35GB) Spec Decode Eagle 1: DeepSeek + Qwen" key: spec-decode-eagle-1-deepseek-qwen - automatic_shard: true timeout_in_minutes: 25 device: h200_35gb source_file_dependencies: @@ -47,7 +44,6 @@ steps: - pytest -v -s v1/e2e/spec_decode/eagle/ -k "deepseek_eagle or qwen3_eagle3" mirror: amd: - automatic_shard: true label: ":amd: (MI355 DPX) Spec Decode Eagle 1: DeepSeek + Qwen" dind: false device: mi355_dpx @@ -65,7 +61,6 @@ steps: - label: ":nvidia: (H200 MIG 35GB) Spec Decode Eagle 2: Llama 3 + Qwen VL + Other" key: spec-decode-eagle-2-llama3-qwen-vl-other - automatic_shard: true timeout_in_minutes: 25 device: h200_35gb source_file_dependencies: @@ -76,7 +71,6 @@ steps: - pytest -v -s v1/e2e/spec_decode/eagle/ -k "not deepseek_eagle and not qwen3_eagle3" mirror: amd: - automatic_shard: true label: ":amd: (MI355 DPX) Spec Decode Eagle 2: Llama 3 + Qwen VL + Other" dind: false device: mi355_dpx @@ -94,7 +88,6 @@ steps: - label: ":nvidia: (B200) [Nightly] Spec Decode Eagle" key: spec-decode-eagle-nightly-b200 - automatic_shard: true timeout_in_minutes: 25 device: b200-k8s optional: true @@ -107,7 +100,6 @@ steps: mirror: amd: label: ":amd: (MI355) Spec Decode Eagle Nightly" - automatic_shard: true dind: false device: mi355_1 num_devices: 1 @@ -131,7 +123,6 @@ steps: # each job below runs whole directories. - label: ":nvidia: (H200 MIG 35GB) Spec Decode Speculators" key: spec-decode-speculators - automatic_shard: true timeout_in_minutes: 30 device: h200_35gb source_file_dependencies: @@ -144,7 +135,6 @@ steps: - pytest -v -s v1/e2e/spec_decode/speculators/ mirror: amd: - automatic_shard: true label: ":amd: (MI355 DPX) Spec Decode Speculators" dind: false device: mi355_dpx @@ -163,7 +153,6 @@ steps: - label: ":nvidia: (H200 MIG 35GB) Spec Decode MTP Gemma4" key: spec-decode-mtp-gemma4 - automatic_shard: true timeout_in_minutes: 30 device: h200_35gb source_file_dependencies: @@ -176,7 +165,6 @@ steps: - pytest -v -s v1/e2e/spec_decode/mtp/gemma4/ mirror: amd: - automatic_shard: true label: ":amd: (MI355 DPX) Spec Decode MTP Gemma4" dind: false device: mi355_dpx @@ -195,7 +183,6 @@ steps: - label: ":nvidia: (H200 MIG 35GB) Spec Decode MTP DeepSeek + MiMo + GLM" key: spec-decode-mtp-deepseek-mimo - automatic_shard: true timeout_in_minutes: 30 device: h200_35gb source_file_dependencies: @@ -213,7 +200,6 @@ steps: - pytest -v -s v1/e2e/spec_decode/mtp/glm/ mirror: amd: - automatic_shard: true label: ":amd: (MI355 DPX) Spec Decode MTP DeepSeek + MiMo + GLM" dind: false device: mi355_dpx @@ -235,7 +221,6 @@ steps: - label: ":nvidia: (H200 MIG 35GB) Spec Decode MTP Qwen3.5" key: spec-decode-mtp-qwen3-5 - automatic_shard: true timeout_in_minutes: 30 device: h200_35gb source_file_dependencies: @@ -252,7 +237,6 @@ steps: - pytest -v -s v1/e2e/spec_decode/mtp/qwen3_5/ mirror: amd: - automatic_shard: true label: ":amd: (MI355 DPX) Spec Decode MTP Qwen3.5" dind: false device: mi355_dpx @@ -274,7 +258,6 @@ steps: - label: ":nvidia: (B200) [Nightly] Spec Decode Speculators + MTP" key: spec-decode-speculators-mtp-nightly-b200 - automatic_shard: true timeout_in_minutes: 30 device: b200-k8s optional: true @@ -289,7 +272,6 @@ steps: mirror: amd: label: ":amd: (MI355) Spec Decode Speculators + MTP Nightly" - automatic_shard: true dind: false device: mi355_1 num_devices: 1 @@ -312,7 +294,6 @@ steps: - label: ":nvidia: (H200 MIG 35GB) Spec Decode N-Gram + Suffix" key: spec-decode-ngram-suffix - automatic_shard: true timeout_in_minutes: 20 device: h200_35gb source_file_dependencies: @@ -325,7 +306,6 @@ steps: - pytest -v -s v1/e2e/spec_decode/ngram_suffix/ mirror: amd: - automatic_shard: true label: ":amd: (MI355 DPX) Spec Decode N-Gram + Suffix" dind: false device: mi355_dpx @@ -345,7 +325,6 @@ steps: - label: ":nvidia: (H200 MIG 18GB) Spec Decode Draft Model" key: spec-decode-draft-model - automatic_shard: true timeout_in_minutes: 60 device: h200_18gb source_file_dependencies: @@ -357,7 +336,6 @@ steps: mirror: amd: label: ":amd: (MI355 DPX) Spec Decode Draft Model" - automatic_shard: true dind: false device: mi355_dpx timeout_in_minutes: 60 @@ -374,7 +352,6 @@ steps: - label: ":nvidia: (B200) [Nightly] Spec Decode Draft Model" key: spec-decode-draft-model-nightly-b200 - automatic_shard: true timeout_in_minutes: 40 device: b200-k8s optional: true @@ -392,7 +369,6 @@ steps: mirror: amd: label: ":amd: (MI355) Spec Decode Draft Model Nightly" - automatic_shard: true dind: false device: mi355_1 num_devices: 1 @@ -414,7 +390,6 @@ steps: - label: ":nvidia: (H100) [Nightly] Speculators Correctness" key: speculators-correctness - automatic_shard: true timeout_in_minutes: 30 device: h100 optional: true @@ -429,7 +404,6 @@ steps: - pytest -v -s v1/spec_decode/test_speculators_correctness.py -m slow_test mirror: amd: - automatic_shard: true label: ":amd: (MI355 DPX) Speculators Correctness Nightly" dind: false device: mi355_dpx @@ -442,7 +416,6 @@ steps: - label: ":nvidia: (B200) [Nightly] Spec Decode DeepSeek MTP Parallel Load" key: spec-decode-deepseek-mtp-parallel-load-2xb200-2xmi300 - automatic_shard: true timeout_in_minutes: 30 device: b200-k8s optional: true @@ -459,7 +432,6 @@ steps: - pytest -v -s v1/e2e/spec_decode/test_mtp_parallel_load.py mirror: amd: - automatic_shard: true label: ":amd: (MI355) Spec Decode DeepSeek MTP Parallel Load" dind: false device: mi355_2 @@ -481,7 +453,6 @@ steps: - label: ":nvidia: (H200 MIG 35GB) [Nightly] Spec Decode AL DFlash" key: spec-decode-dflash-nightly - automatic_shard: true timeout_in_minutes: 90 device: h200_35gb optional: true @@ -502,7 +473,6 @@ steps: mirror: amd: label: ":amd: (MI355 DPX) Spec Decode AL DFlash Nightly" - automatic_shard: true dind: false device: mi355_dpx timeout_in_minutes: 120 @@ -528,7 +498,6 @@ steps: - label: ":nvidia: (B200) [Nightly] Spec Decode AL DFlash2" key: spec-decode-dflash2-nightly-b200 - automatic_shard: true timeout_in_minutes: 60 device: b200-k8s optional: true @@ -545,7 +514,6 @@ steps: - pytest -v -s v1/e2e/spec_decode/acceptance_rates/dflash/ -k "dflash2" mirror: amd: - automatic_shard: true label: ":amd: (MI355) Spec Decode AL DFlash2 Nightly" dind: false device: mi355_1 @@ -575,7 +543,6 @@ steps: - label: ":nvidia: (H100) [Nightly] Spec Decode AL XPress" key: spec-decode-xpress-nightly - automatic_shard: true timeout_in_minutes: 30 device: h100 optional: true @@ -588,7 +555,6 @@ steps: - pytest -v -s v1/e2e/spec_decode/acceptance_rates/xpress/ mirror: amd: - automatic_shard: true label: ":amd: (MI355) Spec Decode AL XPress Nightly" dind: false device: mi355_1 @@ -609,7 +575,6 @@ steps: - label: ":nvidia: (H200 MIG 35GB) [Nightly] Spec Decode AL DSpark" key: spec-decode-dspark-nightly - automatic_shard: true timeout_in_minutes: 60 device: h200_35gb optional: true @@ -624,7 +589,6 @@ steps: mirror: amd: label: ":amd: (MI355 DPX) Spec Decode AL DSpark Nightly" - automatic_shard: true dind: false device: mi355_dpx timeout_in_minutes: 90 @@ -651,7 +615,6 @@ steps: - label: ":nvidia: (H200 MIG 35GB) [Nightly] Spec Decode AL MTP + Other Acceptance" key: spec-decode-mtp-other-acceptance-nightly - automatic_shard: true timeout_in_minutes: 35 device: h200_35gb optional: true @@ -666,7 +629,6 @@ steps: amd: label: ":amd: (MI355 DPX) Spec Decode AL MTP + Other Acceptance Nightly" dind: false - automatic_shard: true device: mi355_dpx timeout_in_minutes: 55 depends_on: diff --git a/.buildkite/test_areas/torch_abi.yaml b/.buildkite/test_areas/torch_abi.yaml index 483c896ffcc7..79bd9bd97289 100644 --- a/.buildkite/test_areas/torch_abi.yaml +++ b/.buildkite/test_areas/torch_abi.yaml @@ -4,7 +4,6 @@ depends_on: steps: - label: ":nvidia: (L4) Torch Stable ABI Audit" key: torch-stable-abi-audit - automatic_shard: true timeout_in_minutes: 10 source_file_dependencies: - .buildkite/check-torch-abi.py