Torchtune development has been discontinued. Have you considered switching to a different multi-GPU train backend?

[ninehills]

ref: https://github.com/meta-pytorch/torchtune/issues/2883

Candidates

• torchforge
• verl
• TRL
• prime-rl

[iqiancheng]

the Candidate Post-training frameworks:

• veOmni
• automodel
• llama-factory
• ms-swift

RL :

• veRL

[Kovbo]

Yes, we are currently experimenting with Megatron to support bigger models. We might also consider other options.