Apple researchers have developed a new method for training large language models (LLMs) that seamlessly integrates both text and visual information.
The company's findings, detailed in a research paper titled "MM1: Methods, Analysis & Insights from Multimodal LLM Pre-training," showcase a new approach to creating more intelligent and flexible AI systems. By utilizing a diverse dataset comprising image-caption pairs, interleaved image-text documents, and text-only data, Apple's claims that the MM1 model sets a new standard in AI's ability to perform tasks such as image captioning, visual question answering, and natural language inference with a high level of accuracy.
Apple's research focuses on the combination of different types of training data and model architectures, which enables the AI to understand and generate language based on a mix of visual and linguistic cues. This capability is vital for tasks that require a nuanced comprehension of the world, such as interpreting complex images or answering questions that involve visual elements.
The paper also highlights the MM1 model's exceptional in-context learning abilities, particularly in the largest 30 billion parameter configuration of the model. This version apparently exhibits remarkable capabilities for multi-step reasoning over multiple images using few-shot "chain-of-thought" prompting, a technique that allows the AI to perform complex, open-ended problem solving based on minimal examples.
This research emerges as part of Apple's broader initiative to enhance its AI capabilities amid growing competition. Earlier today, Bloomberg's Mark Gurman reported that Apple is in discussions with Google to license Google's Gemini generative large-language models to power new features coming to the iPhone as part of iOS 18.
Popular Stories
Apple today announced a "special Apple Experience" in New York, London, and Shanghai, taking place on March 4, 2026 at 9:00am ET.
Apple invited select members of the media to the event in three major cities around the world. It is simply described as a "special Apple Experience," and there is no further information about what it may entail. The invitation features a 3D Apple logo design...
Wednesday February 18, 2026 12:43 pm PST by Joe Rossignol
Apple is looking for a "breakthrough" with its push into wearable AI devices, including an "AirTag-sized pendant," according to Bloomberg's Mark Gurman.
In a report this week, he said the pendant is reminiscent of the failed Humane AI Pin, but it would be an iPhone accessory rather than a standalone product.
The pendant would feature an "always-on" camera and a microphone for Siri voice...
Wednesday February 18, 2026 9:29 am PST by Joe Rossignol
Back at WWDC 2025, Apple revealed that it was planning to allow CarPlay users to watch video via AirPlay in their vehicles while they are not driving, and the first beta of iOS 26.4 suggests the feature may be nearing availability.
There are several new references to CarPlay video streaming functionality within the iOS 26.4 beta's source code. The feature is not yet visible to users, but...
New trade-in data indicates that Apple's iPhone 17 Pro Max has rapidly become the single most traded-in smartphone.
According to a new report from SellCell, Apple's latest flagship iPhone has quickly risen to the top of the independent trade-in market, accounting for 11.5% of all devices appearing in the top-20 trade-in rankings just months after release. The analysis is based on SellCell...
Tuesday February 17, 2026 8:08 am PST by Joe Rossignol
Apple on Monday invited selected journalists and content creators to a "special Apple Experience" on Wednesday, March 4 in New York, London, and Shanghai.
At an Apple Experience, attendees are typically given the opportunity to try out Apple's latest hardware or software. Following the launch of Apple Creator Studio last month, for example, some content creators attended an Apple Experience...