unsloth icon indicating copy to clipboard operation
unsloth copied to clipboard

Llama 3.2 vision support

Open ankitjsrnit opened this issue 1 year ago • 3 comments

Are you going to add llama 3.2 vision model support in near future.

ankitjsrnit avatar Sep 26 '24 23:09 ankitjsrnit

Yes working on it!

danielhanchen avatar Sep 27 '24 00:09 danielhanchen

Can't wait for it, coz it'll be relevant to my curr rsrch.

EvanKA avatar Sep 30 '24 02:09 EvanKA

Hopefully out soon.

danielhanchen avatar Oct 01 '24 05:10 danielhanchen

Tried to run the provided .ipynb on Google Colab Free, changing the based model to Llama 3.2 11B Vision, but it returns an error message as in the attached screenshot. err_msg_llama_3 2_11b_vision_google_colab

EvanKA avatar Oct 08 '24 07:10 EvanKA

Any updates?

yukiarimo avatar Nov 13 '24 08:11 yukiarimo

It's not supp'd yet, but the error msg has been chg'd to reflect that. Screenshot 2024-11-14 at 8 10 31 AM

EvanKA avatar Nov 14 '24 01:11 EvanKA

It's not supp'd yet, but the error msg has been chg'd to reflect that. Screenshot 2024-11-14 at 8 10 31 AM

Do you know how the dataset input format is in 3.2 vision...

geanderson-ai avatar Nov 16 '24 03:11 geanderson-ai

@EvanKA @geanderson-ai @yukiarimo @ankitjsrnit @Nazzaroth2

Hey guys apologies for the delays. Vision models are now supported in Unsloth! Please update Unsloth :)

Read our blogpost: https://unsloth.ai/blog/vision Tweet: https://x.com/UnslothAI/status/1859667930075758793 GitHub post: https://github.com/unslothai/unsloth/releases/tag/November-2024 Vision Model uploads: https://huggingface.co/collections/unsloth/vision-multimodal-models-673eb9908fc2cb3deebd2fa3

shimmyshimmer avatar Nov 22 '24 02:11 shimmyshimmer