Skip to content

[test] add e2e test for qwen3 moe w4a16 #2071

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Draft
wants to merge 5 commits into
base: main
Choose a base branch
from
+30 −0
Draft

[test] add e2e test for qwen3 moe w4a16 #2071

Show file tree
Hide file tree
Changes from 1 commit
Commits
Show all changes
5 commits
Select commit Hold shift + click to select a range
a8180ee
[test] qwen3 moe w4a16 + skip
HDCharles Nov 25, 2025
90c0766
Update qwen3_w4a16_grouped_quant.yaml
HDCharles Dec 1, 2025
a1739ff
update name
HDCharles Dec 1, 2025
ff1180f
ignore gate
HDCharles Dec 1, 2025
39468f2
update save_dir
HDCharles Dec 1, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
9 changes: 9 additions & 0 deletions tests/e2e/vLLM/configs/qwen3_w4a16_grouped_quant.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,9 @@
cadence: "nightly"
test_type: "regression"
model: Qwen/Qwen3-30B-A3B
dataset_id: HuggingFaceH4/ultrachat_200k
dataset_split: train_sft
scheme: W4A16_group
num_calibration_samples: 20
save_dir: "Qwen3-30B-A3B-W4A16-first-10"
recipe: tests/e2e/vLLM/recipes/WNA16/recipe_w4a16_group_quant_first_10_layers.yaml
20 changes: 20 additions & 0 deletions tests/e2e/vLLM/recipes/WNA16/recipe_w4a16_group_quant_first_10_layers.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,20 @@
quant_stage:
quant_modifiers:
GPTQModifier:
ignore: [
"lm_head",
# Ignore layers (10+)
"re:.*model\\.layers\\.([1-9][0-9])\\..*",
]
actorder: null
config_groups:
group_0:
weights:
num_bits: 4
type: "int"
symmetric: True
strategy: "group"
group_size: 128
input_activations: null
output_activations: null
targets: ["Linear"]
Loading