xinyun/vllm
1
0
Fork 0
mirror of https://git.datalinker.icu/vllm-project/vllm.git synced 2025-12-22 02:55:01 +08:00
Code Issues Packages Projects Releases Wiki Activity

[Bugfix] Fix mismatched nvfp4 gemm output shape (#29742)

Browse Source 
Signed-off-by: Isotr0py <mozf@mail2.sysu.edu.cn>
This commit is contained in:
Isotr0py 2025-11-30 17:15:01 +08:00 committed by GitHub
parent 2afcec4dec
commit 47539cfd3e
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
1 changed files with 1 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
vllm/model_executor/layers/quantization/compressed_tensors/schemes/compressed_tensors_w4a4_nvfp4.py
View File

@ -184,7 +184,7 @@ class CompressedTensorsW4A4Fp4(CompressedTensorsScheme):
return out return out
output_dtype = x.dtype output_dtype = x.dtype
output_shape = [x.shape[0], layer.weight_packed.shape[0]] output_shape = [*x.shape[:-1], layer.weight_packed.shape[0]]
# quantize BF16 or FP16 to (FP4 and interleaved block scale) # quantize BF16 or FP16 to (FP4 and interleaved block scale)
x_fp4, x_blockscale = scaled_fp4_quant(x, layer.input_global_scale) x_fp4, x_blockscale = scaled_fp4_quant(x, layer.input_global_scale)