A high-throughput and memory-efficient inference and serving engine for LLMs
Mirrored from https://github.com/tduncan-lum/Nvidia-Optimization-Stack---VLLM.git
Synced Nov 14, 05:30 UTC
[Misc] Remove dead return (#23061)
Signed-off-by: Woosuk Kwon <woosuk.kwon@berkeley.edu>
Woosuk Kwon committed
Aug 17, 2025 at 10:36 UTC
c55bc1db26f5e4385c8a2c1b7e6ba8b54ab2e060
1 file changed
-1
vllm/model_executor/models/qwen2_vl.py
-1
@@ -1225,7 +1225,6 @@ class Qwen2VLForConditionalGeneration(nn.Module, SupportsMultiModal,
1225
modalities = self._parse_and_validate_multimodal_inputs(**kwargs)
1226
if not modalities:
1227
return []
1228
- return None
1228
1229
# The result multimodal_embeddings is tuple of tensors, with each
1230
# tensor correspoending to a multimodal data item (image or video).