Closed j-min closed 7 months ago
I have the same question too!
There should not be any reason for the principles behind SoM to work with any LMM/VLM. This repository does not implement integrations with other models, but in principle it should be straightforward.
Impressive work 👍
I'm interested if SoM works with open-sourced multimodal LMs such as LLaVA v1.5 as well. If you have tried this, can you share your experience on this?