Multi-GPU inference problem.

[jiazheng-xing]

File "/slurm-files/ock/fsar_llava/fsar_llava/videollava/model/llava_arch.py", line 207, in prepare_inputs_labels_for_multimodal
video_features_minibatch = self.encode_videos(videos_minibatch) # fake list [mini_b, t, l, c]
RuntimeError: Expected all tensors to be on the same device, but found at least two devices, cuda:3 and cuda:0! (when checking argument for argument mat2 in method wrapper_CUDA_mm)

[EvanChan8]

Same issue, have you solved it yet？