Multimodality and Local Fine-Tuning

[King-Darius]

Is there any intention of bringing in images and sound using something like SD and say Audiocraft? https://github.com/facebookresearch/audiocraft kind of like how GPT-4 can currently with Dalle-3? Also, what about fine-tuning models within GPT4ALL?

[cebtenzzre]

We are working on image input at least, see #1568.