[CI] Shard Humming A100 eval (#52325)

Signed-off-by: Kevin Luu <51931015+khluu@users.noreply.github.com>
This commit is contained in:
Kevin H. Luu
2026-08-17 02:03:33 -07:00
committed by GitHub
parent c05d923f18
commit bb233626ca
4 changed files with 17 additions and 2 deletions
+3 -2
View File
@@ -195,12 +195,13 @@ steps:
commands:
- pytest -s -v evals/gsm8k/test_gsm8k_correctness.py --config-list-file=evals/gsm8k/configs/moe-refactor-dp-ep/config-b200.txt
- label: LM Eval Humming f16 (A100 - TEMPORARY)
- label: LM Eval Humming f16 (A100 - TEMPORARY) %N
key: lm-eval-humming-f16-a100
timeout_in_minutes: 75
device: a100
optional: true
num_devices: 1
parallelism: 3
source_file_dependencies:
- vllm/model_executor/layers/quantization/humming.py
- vllm/model_executor/layers/quantization/utils/humming_utils.py
@@ -208,7 +209,7 @@ steps:
- vllm/model_executor/layers/fused_moe/oracle/
- vllm/model_executor/kernels/linear/
commands:
- pytest -s -v evals/gsm8k/test_gsm8k_correctness.py --config-list-file=evals/gsm8k/configs/humming/config.txt
- pytest -s -v evals/gsm8k/test_gsm8k_correctness.py --config-list-file=evals/gsm8k/configs/humming/config-a100-shard-$$BUILDKITE_PARALLEL_JOB.txt
- label: LM Eval Humming Act int8 (A100 - TEMPORARY)
key: lm-eval-humming-act-a100
@@ -0,0 +1,4 @@
Qwen3.5-35B-A3B-experts-int8-humming.yaml
gpt-oss-20b-humming.yaml
Qwen3-30B-A3B-GPTQ-Int4-humming.yaml
Qwen3-30B-A3B-FP8-block-humming.yaml
@@ -0,0 +1,5 @@
Qwen3.6-35B-A3B-NVFP4-humming.yaml
Qwen3-30B-A3B-Instruct-2507-quantized.w8a8-humming.yaml
Qwen3-30B-A3B-Fp8-v1-humming.yaml
Qwen3-30B-A3B-AWQ-humming.yaml
Qwen3-0.6B-MXFP8-humming.yaml
@@ -0,0 +1,5 @@
Qwen3.5-35B-A3B-FP8-humming.yaml
NVIDIA-Nemotron-3-Nano-30B-A3B-NVFP4-humming.yaml
Qwen3-30B-A3B-MXFP4A16-humming.yaml
Qwen3-30B-A3B-NVFP4-humming.yaml
Qwen2-1.5B-Instruct-FP8W8-humming.yaml