Join the conversation

Join the community of Machine Learners and AI enthusiasts.

Sign Up
HugoLaurencon 
posted an update May 6
Post
2842
We release Idefics2-chatty, the chatbot-optimized version of Idefics2: HuggingFaceM4/idefics2-8b-chatty

Idefics2-chatty is better at following instructions and following Chain-of-Thoughts reasoning.

Moreover, we also release a paper, containing a lot of findings on how to build an efficient and performant Vision-Language Model: What matters when building vision-language models? (2405.02246)

How are you going to use the model, or what data are you going to fine-tune it on?
This comment has been hidden
This comment has been hidden

Would you share the total training cost info? as traing of IDEFICS2-8B used "approximately 1.5 billion images and 225 billion text tokens" which is quite huge for a 8B sized LMM model

·

It was roughly trained for 1 month on 32 nodes of 8 H100s

error