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 11, 12:30 UTC
[Bugfix] Update get_processor_data to use get_all method (#30583)
Signed-off-by: Drew Botwinick <6953152+dbotwinick@users.noreply.github.com>
Drew Botwinick committed
Dec 14, 2025 at 07:19 UTC
5b64ac21f99ff1c31f5481267ee80e34b3c77955
1 file changed
+1
-1
vllm/multimodal/parse.py
+1
-1
@@ -120,7 +120,7 @@ class ProcessorBatchItems(ModalityDataItems[Sequence[_T], _T]):
120
return self.data[index]
121
122
def get_processor_data(self) -> Mapping[str, object]:
123
- return {f"{self.modality}s": self.data}
123
+ return {f"{self.modality}s": self.get_all()}
124
125
def get_passthrough_data(self) -> Mapping[str, object]:
126
return {}