iHore/vLLM
8
0
Fork 0
Code Issues Pull Requests Actions 1 Packages Projects Releases Wiki Activity

[CI/Build] Fix test_deepseek_v41_mxfp8_scale_loading after #57797 (#61678)

Browse Source 
Signed-off-by: Stefano Castagnetta <scastagnetta@nvidia.com>
Co-authored-by: Misha Goin <mgoin64@gmail.com>
This commit is contained in:
Stefano Castagnetta
2026-10-14 12:17:59 -07:00
committed by GitHub
co-authored by Misha Goin
parent e96c822660
commit 222e00fc44
1 changed files with 3 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files
tests/quantization/test_fp8.py
+3 -1
View File
@@ -79,7 +79,9 @@ def test_deepseek_v41_mxfp8_scale_loading(
quant_config._resolved_expert_dtype = "fp4"
for module in (model_module, linear_module):
monkeypatch.setattr(module, "get_tensor_model_parallel_world_size", lambda: 2)
monkeypatch.setattr(module, "get_tensor_model_parallel_rank", lambda: tp_rank)
monkeypatch.setattr(
linear_module, "get_tensor_model_parallel_rank", lambda: tp_rank
)
block = torch.nn.Module()
block.attn = torch.nn.Module()
block.attn.wq_b = ColumnParallelLinear(