biondizzle/vllm
1
0
Fork 0
Code Issues Pull Requests Actions 2 Packages Projects Releases Wiki Activity

[NVIDIA] Bugfix NVFP4 DGX Spark and RTX50 (#38423)

Browse Source 
Signed-off-by: johnnynunez <johnnynuca14@gmail.com>
Signed-off-by: Johnny <johnnynuca14@gmail.com>
This commit is contained in:
Johnny
2026-03-30 18:36:18 +02:00
committed by GitHub
parent 8e6293e838
commit b4a2f3ac36
15 changed files with 86 additions and 20 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
tests/kernels/moe/test_unquantized_backend_selection.py
View File

@@ -57,7 +57,6 @@ def test_select_default_backend_by_platform(
moe_config = make_dummy_moe_config()
selected_backend = select_unquantized_moe_backend(
moe_config=moe_config,
use_ep=False,
use_dp=False,
)
@@ -90,7 +89,6 @@ def test_select_rocm_aiter_backend(mock_aiter_enabled, mock_has_flashinfer):
moe_config = make_dummy_moe_config()
selected_backend = select_unquantized_moe_backend(
moe_config=moe_config,
use_ep=False,
use_dp=False,
)
@@ -129,7 +127,6 @@ def test_select_cuda_flashinfer_trtllm_backend(
selected_backend = select_unquantized_moe_backend(
moe_config=moe_config,
use_ep=True,
use_dp=False,
)
@@ -171,7 +168,6 @@ def test_select_cuda_flashinfer_cutlass_backend(
selected_backend = select_unquantized_moe_backend(
moe_config=moe_config,
use_ep=True, # CUTLASS requires EP
use_dp=False, # CUTLASS doesn't support DP
)