Skip to content

Adding new MoE e2e tests #1960

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

Open
wants to merge 9 commits into
base: main
Choose a base branch
from
+12 −1
Open

Adding new MoE e2e tests #1960

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
9 commits
Select commit Hold shift + click to select a range
ca325c7
Adding new MoE e2e tests [wip]
HDCharles Oct 22, 2025
2289cdf
making tests run faster
HDCharles Nov 4, 2025
e42680f
further shaving int4 layers to improve e2e test time
HDCharles Nov 6, 2025
565f676
formatting
HDCharles Nov 6, 2025
6d6f237
fix error
HDCharles Nov 6, 2025
a7e1a90
needs to be symmetric for MoE
HDCharles Nov 7, 2025
5f466d9
removing int4 test
HDCharles Nov 11, 2025
709dc98
Merge branch 'main' into 02_moe_e2e
HDCharles Nov 11, 2025
581a53a
Merge branch 'main' into 02_moe_e2e
HDCharles Nov 12, 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
7 changes: 7 additions & 0 deletions tests/e2e/vLLM/configs/qwen3_fp4_nvfp4.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,7 @@
cadence: "nightly"
test_type: "regression"
model: Qwen/Qwen3-30B-A3B
scheme: NVFP4
dataset_id: HuggingFaceH4/ultrachat_200k
dataset_split: train_sft
num_calibration_samples: 20
4 changes: 4 additions & 0 deletions tests/e2e/vLLM/configs/qwen3_fp8_dynamic_per_token.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,4 @@
cadence: "nightly"
test_type: "regression"
model: Qwen/Qwen3-30B-A3B
scheme: FP8_DYNAMIC
2 changes: 1 addition & 1 deletion tests/e2e/vLLM/run_vllm.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -18,7 +18,7 @@ def parse_args():
except json.JSONDecodeError as e:
raise ValueError(f"Invalid JSON input: {e}")

if "W4A16_2of4" in scheme:
if scheme is not None and "W4A16_2of4" in scheme:
# required by the kernel
llm_kwargs["dtype"] = torch.float16

Expand Down