[BugFix] Fixes Qwen3-Next enable nz accuracy problem (#4058)

### What this PR does / why we need it?
- Fixes Qwen3-Next enable nz accuracy problem

### Does this PR introduce _any_ user-facing change?
N/A


- vLLM version: v0.11.0
- vLLM main:
83f478bb19

---------

Signed-off-by: Icey <1790571317@qq.com>
Signed-off-by: wxsIcey <1790571317@qq.com>
This commit is contained in:
Icey
2025-11-10 20:54:57 +08:00
committed by GitHub
parent e6625bb582
commit e04a87f4be
8 changed files with 36 additions and 20 deletions

View File

@@ -416,6 +416,7 @@ class TestAscendMLAImpl(TestBase):
self.assertEqual(q_pe.shape[1], self.impl.num_heads)
self.assertEqual(q_pe.shape[2], self.impl.qk_rope_head_dim)
@patch('vllm_ascend.utils._ENABLE_NZ', True)
@patch('torch_npu.npu_format_cast')
def test_process_weights_after_loading(self, mock_format_cast):
layer = MagicMock(spec=LinearBase)