VLLM plugin and Xinference plugin result in GPU memory overflow #34

Open
opened 2026-02-22 17:16:39 -05:00 by yindo · 0 comments
Owner

Originally created by @moshilangzi on GitHub (May 9, 2025).

The VLLM plugin and Xinference plugin exhibit the same problem. Directly calling the VLLM and Xinference APIs to upload multiple images and perform multi-image question answering works correctly. However, when uploading multiple images using the VLLM and Xinference plugins on the Dify platform, it results in GPU memory overflow.

Originally created by @moshilangzi on GitHub (May 9, 2025). The VLLM plugin and Xinference plugin exhibit the same problem. Directly calling the VLLM and Xinference APIs to upload multiple images and perform multi-image question answering works correctly. However, when uploading multiple images using the VLLM and Xinference plugins on the Dify platform, it results in GPU memory overflow.
Sign in to join this conversation.
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: langgenius/dify-plugins#34