group: LM Eval
depends_on: 
  - image-build
steps:
- label: LM Eval Small Models
  timeout_in_minutes: 75
  source_file_dependencies:
  - csrc/
  - vllm/model_executor/layers/quantization
  autorun_on_main: true
  commands:
  - pytest -s -v evals/gsm8k/test_gsm8k_correctness.py --config-list-file=configs/models-small.txt

- label: LM Eval Large Models (4 GPUs)(A100)
  device: a100
  optional: true
  num_devices: 4
  working_dir: "/vllm-workspace/.buildkite/lm-eval-harness"
  source_file_dependencies:
  - csrc/
  - vllm/model_executor/layers/quantization
  commands:
  - export VLLM_WORKER_MULTIPROC_METHOD=spawn
  - pytest -s -v test_lm_eval_correctness.py --config-list-file=configs/models-large.txt --tp-size=4

- label: LM Eval Large Models (4 GPUs)(H100)
  device: h100
  optional: true
  num_devices: 4
  working_dir: "/vllm-workspace/.buildkite/lm-eval-harness"
  source_file_dependencies:
  - csrc/
  - vllm/model_executor/layers/quantization
  commands:
    - export VLLM_USE_DEEP_GEMM=0  # We found Triton is faster than DeepGEMM for H100
    - pytest -s -v test_lm_eval_correctness.py --config-list-file=configs/models-large-hopper.txt --tp-size=4

- label: LM Eval Small Models (B200)
  timeout_in_minutes: 120
  device: b200
  optional: true
  source_file_dependencies:
  - csrc/
  - vllm/model_executor/layers/quantization
  commands:
  - pytest -s -v evals/gsm8k/test_gsm8k_correctness.py --config-list-file=configs/models-blackwell.txt

- label: LM Eval Large Models (H200)
  timeout_in_minutes: 60
  device: h200
  optional: true
  num_devices: 8
  commands:
    - pytest -s -v evals/gsm8k/test_gsm8k_correctness.py --config-list-file=configs/models-h200.txt

- label: MoE Refactor Integration Test (H100 - TEMPORARY)
  device: h100
  optional: true
  num_devices: 2
  commands:
    - pytest -s -v evals/gsm8k/test_gsm8k_correctness.py --config-list-file=evals/gsm8k/configs/moe-refactor/config-h100.txt
  
- label: MoE Refactor Integration Test (B200 - TEMPORARY)
  device: b200
  optional: true
  num_devices: 2
  commands:
    - pytest -s -v evals/gsm8k/test_gsm8k_correctness.py --config-list-file=evals/gsm8k/configs/moe-refactor/config-b200.txt

- label: MoE Refactor Integration Test (B200 DP - TEMPORARY)
  device: b200
  optional: true
  num_devices: 2
  commands:
    - pytest -s -v evals/gsm8k/test_gsm8k_correctness.py --config-list-file=evals/gsm8k/configs/moe-refactor-dp-ep/config-b200.txt

- label: GPQA Eval (GPT-OSS) (H100)
  timeout_in_minutes: 120
  device: h100
  optional: true
  num_devices: 2
  source_file_dependencies:
  - csrc/
  - vllm/model_executor/layers/quantization
  - tests/evals/gpt_oss/
  commands:
    - uv pip install --system 'gpt-oss[eval]==0.0.5'
    - pytest -s -v evals/gpt_oss/test_gpqa_correctness.py --config-list-file=configs/models-h100.txt

- label: GPQA Eval (GPT-OSS) (B200)
  timeout_in_minutes: 120
  device: b200
  optional: true
  num_devices: 2
  source_file_dependencies:
  - csrc/
  - vllm/model_executor/layers/quantization
  - tests/evals/gpt_oss/
  commands:
    - uv pip install --system 'gpt-oss[eval]==0.0.5'
    - pytest -s -v evals/gpt_oss/test_gpqa_correctness.py --config-list-file=configs/models-b200.txt