[c0n5pir4cy]

The Yi models were actually released back in early November 2023. So there isn't as big a gap in time as it seems.

I'm not sure why there is such a big gap between the release of the models and the publication of the paper.

EDIT: Okay, this appears to be a new set of models with the same name, based on the same models from November but now with multimodal capabilities.