will to add qwen-vl ? Thanks

[FDInSky]

No description provided.

[LZHgrla]

Now, we are woking on the reproduction of Llava and minigpt, (including Pre-training and fine-tuning!).
Thus, other multi-modal LLMs have not been considered yet.

However, we greatly welcome any ideas of the designing of Qwen-VL fine-tuning, such as "how to use open-source dataset formats to fine-tune Qwen-VL", "which down-stream datasets can be used to fine-tune Qwen-VL", and so on. If anyone have some thoughts, please let us know!

[FDInSky]

Now, we are woking on the reproduction of Llava and minigpt, (including Pre-training and fine-tuning!). Thus, other multi-modal LLMs have not been considered yet.

However, we greatly welcome any ideas of the designing of Qwen-VL fine-tuning, such as "how to use open-source dataset formats to fine-tune Qwen-VL", "which down-stream datasets can be used to fine-tune Qwen-VL", and so on. If anyone have some thoughts, please let us know!

When will open llava or minigpt4，we can follow them to add qwen-vl.

[LZHgrla]

@FDInSky The support of Llava is expected within about one week (at most two weeks :)), and as for the minigpt-4, we needs more time.