Hugging Face's logo

Join the conversation

Join the community of Machine Learners and AI enthusiasts.

Sign Up
Back to feed
xianbao 
posted an update Feb 8
Post
Welcome Bunny! A family of lightweight but powerful multimodal models from BAAI

With detailed work on dataset curation, the Bunny-3B model built upon SigLIP and Phi-2 achieves performance on par with 13B models.

Model: BAAI/bunny-phi-2-siglip-lora

merve

What is the limitation that you haven't use your own EVACLIP here?

·
Isaachhe

Thanks for your attention!
We have tried two vision encoders and three LLMs. The best performances are achieved by integrating SigLIP-SO and Phi-2. We released all weights of combinations. Please refer to our GitHub for more information.

In this post