Apple Came Up With an Open-Source AI Image Editing Model

Open-Source AI Image Editing Model

Apple is venturing further into the realm of artificial intelligence with its recent release of an open-source AI image-editing model, sparking curiosity about the company’s future endeavours in this space. Dubbed “Apple GPT” or MLLM-Guided Image Editing (MGIE), this multimodal AI model represents a significant leap in image editing technology. Developed by researchers from Apple and the University of California, Santa Barbara, MGIE allows users to edit images using simple text commands, akin to the functionalities of popular software like Photoshop.

While Apple has traditionally been tight-lipped about its AI initiatives, the unveiling of MGIE suggests a deeper foray into this domain. Despite refraining from major AI announcements during last year’s ChatGPT craze, Apple CEO Tim Cook hinted at forthcoming developments in the AI landscape.

MGIE stands out from existing AI image editing tools due to its ability to comprehend and execute nuanced text prompts. Traditional methods often falter when interpreting concise instructions, leading to subpar results. In contrast, MGIE leverages multimodal large language models (MLLMs) to decipher text commands and enhance images accordingly.

For instance, MGIE can transform a pepperoni pizza into a healthier version by adding vegetables, based on the command “make this more healthy.” In another example, it accurately incorporates lightning reflections in water, showcasing its advanced capabilities compared to other models.

The availability of MGIE as an open-source model on GitHub and a demo version on Hugging Face signifies Apple’s commitment to fostering innovation and collaboration within the AI community.

As Apple continues to push the boundaries of AI technology, the release of MGIE hints at exciting possibilities for future advancements in image editing and beyond.