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

[Misc] Fix ScalarType float4 naming (#17690)

Browse Source 
Signed-off-by: Lucas Wilkinson <lwilkinson@neuralmagic.com>
This commit is contained in:
Lucas Wilkinson
2025-05-06 04:07:15 -04:00
committed by GitHub
parent 63ced7b43f
commit 6eae34533a
3 changed files with 3 additions and 3 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
tests/kernels/quantization/test_nvfp4_quant.py
View File

@@ -17,7 +17,7 @@ PAD_SHAPES = [(90, 64), (150, 64), (128, 48), (128, 80), (150, 80), (90, 48),
SEEDS = [42]
CUDA_DEVICES = ['cuda:0']
FLOAT4_E2M1_MAX = scalar_types.float4_e2m1fn.max()
FLOAT4_E2M1_MAX = scalar_types.float4_e2m1f.max()
FLOAT8_E4M3_MAX = torch.finfo(torch.float8_e4m3fn).max
# E2M1 to float