xinyun/vllm
1
0
Fork 0
mirror of https://git.datalinker.icu/vllm-project/vllm.git synced 2025-12-14 05:45:01 +08:00
Code Issues Packages Projects Releases Wiki Activity

static_scaled_fp8_quant should not run when scale.numel is not 1 (#20076)

Browse Source
This commit is contained in:
Eldar Kurtić 2025-06-25 21:08:03 +02:00 committed by GitHub
parent 23a04e0895
commit 8b8c209e35
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/_custom_ops.py
View File

@ -1276,7 +1276,7 @@ def scaled_fp8_quant(
torch.ops._C.dynamic_scaled_fp8_quant(output, input, scale) torch.ops._C.dynamic_scaled_fp8_quant(output, input, scale)
else: else:
# num_token_padding not implemented for this case # num_token_padding not implemented for this case
assert (scale.numel() == 1 or num_token_padding is None) assert (scale.numel() == 1 and num_token_padding is None)
torch.ops._C.static_scaled_fp8_quant(output, input, scale) torch.ops._C.static_scaled_fp8_quant(output, input, scale)
return output, scale return output, scale