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 10, 00:15 UTC
[BugFix] Fix embed_input_ids argument error of QwenVLForConditionalGeneration (#32462)
honglyua committed
Jan 19, 2026 at 11:06 UTC
976af2f314d1c7d54e3707e325df2b5961c069a4
1 file changed
+2
vllm/model_executor/models/qwen_vl.py
+2
@@ -731,6 +731,8 @@ class QwenVLForConditionalGeneration(
731
],
732
}
733
734
+ embed_input_ids = SupportsMultiModal.embed_input_ids
735
+
736
def get_mm_mapping(self) -> MultiModelKeys:
737
"""
738
Get the module prefix in multimodal models