Apple Unveils Powerful New MM1 AI Model for Multimodal Understanding
• Apple researchers develop new MM1 AI model that integrates text and visual data for improved language understanding and generation • Model sets new standard in accuracy for image captioning, visual QA, and other multimodal tasks • Largest 30B parameter version shows exceptional few-shot reasoning abilities over images • Research part of broader Apple initiative to enhance AI capabilities amid growing competition • Apple reportedly in talks with Google to license Gemini models to power new iOS 18 iPhone features