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, 04:15 UTC
[Misc] Raise error for missing video metadata in `MultiModalDataParser` (#27664)
Signed-off-by: Isotr0py <mozf@mail2.sysu.edu.cn> Signed-off-by: Isotr0py <2037008807@qq.com> Co-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com>
Isotr0py committed
Oct 29, 2025 at 18:06 UTC
1a33aacf8212472d5c218e3e7b1e8c0f8729cece
1 file changed
+5
vllm/multimodal/parse.py
+5
@@ -506,6 +506,11 @@ class MultiModalDataParser:
506
for data_item in data_items:
507
video, metadata = self._get_video_with_metadata(data_item)
508
if self.video_needs_metadata:
509
+ if metadata is None:
510
+ raise ValueError(
511
+ "Video metadata is required but not found in mm input. "
512
+ "Please check your video input in `multi_modal_data`"
513
+ )
514
new_videos.append((video, metadata))
515
metadata_lst.append(metadata)
516
else: