Apple has published papers on small LLM models and multimodal models already. I would be surprised if they aren't using them for on-device processing.
Apple has published papers on small LLM models and multimodal models already. I would be surprised if they aren't using them for on-device processing.