LLaVA-NeXT About pretrain script and data of LLaVA-NeXT

Hello, I carefully reviewed the content included in the README of this repository and noticed that the LLaVA-NeXT released in 202401 still has not opened the pre-training code and the corresponding dataset, right?

Dec 17 '24 09:12 ligeng0197

I think so. they not open the pre-training code

Jan 27 '25 16:01 xingdi1990

Hello, is the pretraining code/data available ? at least the code for the last phase ?

Feb 11 '25 23:02 Mohamed-Dhouib

data is available here https://huggingface.co/datasets/lmms-lab/LLaVA-NeXT-Data, as for training code, you can use https://github.com/xiaoachen98/Open-LLaVA-NeXT, code is easy to use

Feb 25 '25 08:02 Cooperx521

data is available here https://huggingface.co/datasets/lmms-lab/LLaVA-NeXT-Data, as for training code, you can use https://github.com/xiaoachen98/Open-LLaVA-NeXT, code is easy to use

Hello, thanks for the answer. Sadly, the GitHub given does not provide code to reproduce LLAVA-OneVision training.

Feb 27 '25 18:02 Mohamed-Dhouib