Base PR: https://github.com/ggerganov/llama.cpp/pull/5267   
Note: integration in llama.cpp is partial at this point - it uses llava-1.5 preprocessing  without applying that PR

Important: Verify that processing a simple question with any image at least uses 1200 tokens of prompt processing, that shows that the new PR is in use.
If your prompt is just 576 + a few tokens, you are using llava-1.5 code (or projector) and this is incompatible with llava-1.6



The mmproj files are the embedded ViT's that came with llava-1.6, I've not compared them but given the previous releases from the team I'd be surprised if the ViT has not been fine tuned this time.
If that's the case, using another ViT can cause issues.