RuntimeError: aten.embedding.default: got mixed torch.Tensor and DTensor

[AIprogrammer]

System Info

main dependencies versions are as follows:
torch2.6.0
transformers 4.51.3

Information

• The official example scripts
• My own modified scripts

Tasks

• An officially supported task in the examples folder (such as GLUE/SQuAD, ...)
• My own task or dataset (give details below)

Reproduction

use hf_rollout as rollout method, while running generate_sequences，the error occurs.
'multi_modal_inputs' including input_ids, pixel_values, is stored in non_tensor_batch.

Does anyone know how to solve the discrepancy between the tensor type of inputs and models?

Expected behavior

same as above