Skip to content

[None][test] Add MNNVL to unittest and speed up weight creation #7071

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

Closed
wants to merge 4 commits into from
Closed

[None][test] Add MNNVL to unittest and speed up weight creation #7071

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
4 commits
Select commit Hold shift + click to select a range
3bbcceb
Add MNNVL to unittest and speed up weight creation
yuantailing Aug 20, 2025
a6552df
Merge branch 'main' into unittest_mnnvl
yuantailing Aug 24, 2025
329d08d
Merge branch 'main' into unittest_mnnvl
yuantailing Aug 25, 2025
ba13e99
Merge branch 'main' into unittest_mnnvl
yuantailing Aug 25, 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
1 change: 1 addition & 0 deletions tests/integration/test_lists/test-db/l0_dgx_b200.yml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -15,6 +15,7 @@ l0_dgx_b200:
backend: pytorch
tests:
- unittest/_torch/multi_gpu_modeling -k "deepseek"
- unittest/_torch/modules/test_fused_moe.py::test_fused_moe_alltoall_fp4[MNNVL]
- unittest/_torch/modules/test_fused_moe.py::test_fused_moe_alltoall_fp4[DeepEPLowLatency]
- accuracy/test_llm_api_pytorch.py::TestLlama3_1_8BInstruct::test_bfloat16_4gpus[pp4-attn_backend=TRTLLM-torch_compile=False]
- accuracy/test_llm_api_pytorch.py::TestLlama3_1_8BInstruct::test_fp8_4gpus[tp4-fp8kv=True-attn_backend=TRTLLM-torch_compile=False]
Expand Down
1 change: 1 addition & 0 deletions tests/integration/test_lists/test-db/l0_dgx_h100.yml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -78,6 +78,7 @@ l0_dgx_h100:
auto_trigger: deepseek
tests:
- unittest/_torch/multi_gpu_modeling -k "deepseek"
- unittest/_torch/modules/test_fused_moe.py::test_fused_moe_alltoall[MNNVL]
- unittest/_torch/modules/test_fused_moe.py::test_fused_moe_alltoall[DeepEP]
- unittest/_torch/modules/test_fused_moe.py::test_fused_moe_alltoall[DeepEPLowLatency]
- unittest/_torch/modules/test_fused_moe.py::test_fused_moe_w4afp8[MoEWeightLoadingMode.VANILLA-dtype0]
Expand Down
9 changes: 6 additions & 3 deletions tests/unittest/_torch/modules/test_fused_moe.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -212,11 +212,14 @@ def per_rank_test_fused_moe_alltoall(job_id):
weights = {}
for expert_id in range(NUM_EXPERTS):
w1_weight = torch.empty((INTERMEDIATE_SIZE, HIDDEN_SIZE),
dtype=dtype)
dtype=dtype,
device="cuda")
w2_weight = torch.empty((HIDDEN_SIZE, INTERMEDIATE_SIZE),
dtype=dtype)
dtype=dtype,
device="cuda")
w3_weight = torch.empty((INTERMEDIATE_SIZE, HIDDEN_SIZE),
dtype=dtype)
dtype=dtype,
device="cuda")
torch.nn.init.xavier_uniform_(w1_weight)
torch.nn.init.xavier_uniform_(w2_weight)
torch.nn.init.xavier_uniform_(w3_weight)
Expand Down