# Question about the support of DistributedDataParallel

[mexiQQ]

Hi Jiant team:

Thank you very much for your contribution to the NLP community.

Here I have a question about the distributed training of Jiant and hope to get a response.

I really saw you guys have supported the DDP for the model, however, you do not support the DistributedSampler. Therefore, I doubt will the DDP can accelerate the training process?

And is there any reason that you guys design this paradigm of data providers？

Thx.