Apple's push towards Siri improvements continues with a text-powered open-source image editing AI model

Apple has released MGIE, an open-source multimodal large language model capable of performing complex image edits via text instructions. This tool translates natural language commands into pixel-level adjustments, handling tasks ranging from basic cropping and filtering to advanced operations like object removal, background changes, and artistic style transfers. By making this technology publicly available, Apple provides developers and researchers with a glimpse into the sophisticated AI infrastructure currently under development for its consumer products. The primary implication is that Siri may soon evolve from a voice-activated assistant into a powerful creative tool. Users could potentially instruct Siri to modify photos in their library using natural language, bypassing complex menus in the Photos app. This integration could significantly streamline workflows, allowing for immediate visual alterations within shortautomations or directly through voice commands, thereby bridging the gap between conversational AI and practical image manipulation. This release is highly relevant to open_data because it demonstrates Apple’s strategy of contributing valuable, functional AI assets to the public domain. By sharing MGIE, Apple encourages community experimentation and validation of its underlying models, fostering a healthier open-source ecosystem. This transparency not only accelerates innovation in generative AI but also sets a benchmark for how major technology companies can responsibly share foundational tools, promoting interoperability and broader accessibility in AI-driven creative applications.

Source: imore.com
Published on 2024-02-08

Apple releases open-source AI model that can edit images from natural language instructions »

Open Data

Related news