Apple has made a technological breakthrough: it is expected to run large language models on the iPhone.
Apple’s artificial intelligence researchers said they had achieved a major breakthrough and successfully deployed a large language model (LLM) on Apple devices with limited memory through an innovative flash memory utilization technology. ). This achievement is expected to bring more powerful Siri, real-time language translation, and cutting-edge AI features integrated with photography and augmented reality to future iPhones.
In recent years, LLM chatbots like ChatGPT and Claude have become popular around the world. They can conduct fluent conversations, write text in different styles, and even generate code, demonstrating powerful language understanding and generation capabilities. However, these models have a fatal weakness: they “eat” data and memory so much that ordinary mobile phones simply cannot meet their operating needs.
In order to break through this bottleneck, Apple researchers have opened up a new path and turned their attention to the ubiquitous flash memory in mobile phones, which is where applications and photos are stored. In a paper titled “LLM in a flash: Efficient Large Language Model Inference with Limited Memory,” researchers propose a clever flash memory utilization technique , stores the data of the LLM model in flash memory . The authors point out that flash memory in mobile devices is much larger than the RAM traditionally used to run LLM.
Naijatechnews notes that their approach cleverly utilizes two key technologies to bypass limitations, thereby minimizing data transfer and maximizing flash memory throughput:
-
Windowing: Think of it as a method of recycling. Instead of loading new data every time, the AI model reuses some of the processed data. This reduces the need for frequent memory reads, making the entire process smoother and more efficient.
-
Row-Column Bundling: This technique is similar to when you read a book, instead of reading it word for word, you read it in paragraphs. By grouping data more efficiently, data can be read from flash memory faster, accelerating AI’s ability to understand and generate language.
The paper points out that this technology can allow AI models to run at twice the size of the iPhone’s available memory. With the support of this technology, LLM’s inference speed is increased by 4-5 times on the Apple M1 Max CPU and 20-25 times on the GPU. ” This breakthrough is critical for deploying advanced LLMs in resource-limited environments , greatly expanding their applicability and accessibility,” the researchers wrote.
Faster and more powerful iPhone AI
Breakthroughs in AI efficiency open up new possibilities for future iPhones, such as more advanced Siri functionality, real-time language translation, and sophisticated AI-driven features in photography and augmented reality. The technology also lays the foundation for the iPhone to run sophisticated artificial intelligence assistants and chatbots on the device, something Apple is said to have already begun working on.
The generative AI developed by Apple may eventually be integrated into its Siri voice assistant . In February 2023, Apple held an artificial intelligence summit and introduced its large-scale language model work to employees. According to Bloomberg, Apple’s goal is to create a smarter Siri that is deeply integrated with AI. Apple plans to update the way Siri interacts with the Messages app, allowing users to more efficiently handle complex questions and automatically complete sentences. In addition, Apple is rumored to be planning to add AI to as many Apple apps as possible.
Apple GPT: A super brain in your pocket called Ajax
According to reports, Apple is developing its own generative AI model, codenamed “Ajax”, aiming to compete with OpenAI’s GPT-3 and GPT-4, with 200 billion parameters, implying that it is highly complex in language understanding and generation. sex and powerful abilities. Ajax, known internally as “Apple GPT ,” is designed to unify Apple’s machine learning development, signaling Apple’s deeper integration of AI into its ecosystem.
According to the latest reports, Ajax is considered to be more powerful than the early ChatGPT 3.5. However, some sources also noted that OpenAI’s new model may have exceeded Ajax’s capabilities.
Both The Information and analyst Jeff Pu claim that Apple will offer some form of generative AI capabilities on iPhone and iPad around the end of 2024, when iOS 18 is released. Pu said in October that Apple would build hundreds of AI servers in 2023 and more servers in 2024. According to reports, Apple will provide a solution that combines cloud AI and device-side processing AI.
For more such interesting article like this, app/softwares, games, Gadget Reviews, comparisons, troubleshooting guides, listicles, and tips & tricks related to Windows, Android, iOS, and macOS, follow us on Google News, Facebook, Instagram, Twitter, YouTube, and Pinterest.