New APPLE AI Multimodal Surpassing GPT-4’s Vision

Apple has taken a significant leap in the field of artificial intelligence with the introduction of its new multimodal AI system, Feret. This development marks a pivotal moment as Feret surpasses GPT-4’s capabilities in certain aspects, signaling a shift in the AI leadership landscape.

The Feret Model: A New Benchmark in AI Vision

The Feret model, developed by Apple researchers, is primarily a vision model that has shown impressive results in image identification. It consists of several components:

  1. Clip VIT L14: This tool helps the model understand images by converting them into a format that the computer can process.
  2. Word Processing: It also converts words into a format it can understand.
  3. Precise Identification: The model is adept at identifying specific areas in an image. If a user references a particular part of the picture, like a cat in the bottom-left corner, Feret uses special coordinates to locate it accurately.
  4. Handling Shapes: It is proficient in handling various shapes in the picture, understanding details and locations of each point.
  5. Integration of Information: Feret combines this information to accurately find and describe the specified part of the picture.

Benchmark Comparisons: Feret vs. GPT-4

Feret has demonstrated superior capabilities in specific benchmarks compared to GPT-4. It excels in understanding the relationship between objects in an image and their real-world applications. In tests, Feret showed an advanced ability to pinpoint small and specific regions, outperforming GPT-4 in detailed image analysis.

YouTube video

Practical Applications and Implications

The practical implications of Feret are vast. For instance, its advanced image identification could significantly aid in autonomous driving, potentially surpassing current AI systems in cars. This suggests a future where AI models like Feret could play a critical role in interpreting complex, out-of-context scenarios on the road.

Apple’s Strategic Moves in AI

Apple’s introduction of Feret aligns with its recent pattern of acquiring various AI companies. These acquisitions have enabled Apple to enhance its AI capabilities across its product range, from facial recognition in iPhones to Siri’s improved natural language processing. Moreover, Apple’s commitment to machine learning research and development is evident in its regular publication of innovative research papers.

The Future: Apple GPT and Beyond

Apple’s venture into generative AI with the rumored Apple GPT is another step in this direction. Expected to enhance Siri’s capabilities, Apple GPT could bring about better natural language understanding, improved text generation, and enhanced conversational abilities. This development signifies Apple’s acknowledgment of the rapid advancements in AI and its commitment to not being left behind in this transformative era.

Conclusion: Apple’s Bold Leap into Advanced AI

With Feret, Apple has not only demonstrated its technological prowess but also its strategic foresight in the AI domain. The company is making significant strides in machine learning, challenging existing leaders in the AI space, and shaping the future of how we interact with technology. As we await further developments and the rumored release of Apple GPT, one thing is clear: Apple is a formidable player in the ongoing AI race.