Skip to content

[BUG FIX] Try to enable cuda graph ut #3192

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

Merged
yaox12 merged 1 commit into from
Feb 2, 2026
+0 −6
Merged

[BUG FIX] Try to enable cuda graph ut #3192

Changes from all commits
Commits
Show all changes
1 commit
Select commit
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
6 changes: 0 additions & 6 deletions tests/unit_tests/transformer/test_cuda_graphs.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1252,10 +1252,6 @@ def test_moe_partial_cudagraph(self, ep_size, moe_dropless_dispatcher, moe_dispa
extra_kwargs["moe_token_dispatcher_type"] = "flex"
extra_kwargs["moe_flex_dispatcher_backend"] = "deepep"
elif moe_dispatcher_type == "hybridep":
pytest.skip(
"Currently, the Hybrid EP is broken. "
"Temporarily skip the test and wait for the fix."
)
if not is_hybrid_ep_available():
pytest.skip("Hybrid EP is not available")
extra_kwargs["moe_token_dispatcher_type"] = "flex"
Expand All @@ -1265,8 +1261,6 @@ def test_moe_partial_cudagraph(self, ep_size, moe_dropless_dispatcher, moe_dispa
if not moe_dropless_dispatcher:
if moe_dispatcher_type == "deepep":
pytest.skip("Deep EP doesn't support drop&pad MoE")
if moe_dispatcher_type == "hybridep" and ep_size == 1:
pytest.skip("Hybrid EP doesn't support drop&pad MoE with ep_size == 1")
extra_kwargs["moe_expert_capacity_factor"] = 1.0
extra_kwargs["moe_pad_expert_input_to_capacity"] = True

Expand Down
Loading