Bloomberg reported that Apple is actively negotiating with Google to seek authorization to introduce the Gemini-generated artificial intelligence model into the new iPhone software features this year. At the same time, Apple is also in talks with OpenAI, which provides support for Microsoft's AI capabilities. According to reports, Apple has shown a strong interest in cloud-based generative AI collaboration, with text and image generation mentioned as potential applications for Gemini. However, Apple is also working on developing its own AI models and features for its devices, which are expected to debut in the upcoming iOS 18 release. Currently, discussions between the two parties are still ongoing, and the naming of the AI agreement is yet to be known. This potential collaboration would be a significant expansion of the existing relationship between the two companies (default search engine). Looking at other industry developments, Google has previously announced a partnership with Samsung to provide Gemini-related features for the Galaxy S24 series, such as battery summary for note-taking and recording applications, as well as the generation and editing of text-to-image diffusion. These features rely on server-side processing, but Samsung also uses the device-side version of Gemini. Google offers three sizes of Gemini models, with the Pro version being the most popular among first-party and third-party applications. Gemini 1.0 Pro supports the free version of Gemini, while 1.0 Ultra serves the paid Gemini Advanced layer. It is worth mentioning that the stable version of Gemini 1.0 has been released, and Google also previewed the Gemini 1.5 version with a larger context window in mid-February, which can absorb more information to improve the coherence, relevance, and practicality of the output. In the Gemini series, the Ultra version is positioned as the largest and most powerful model, suitable for highly complex tasks; the Pro version is the best choice for scalable multitasking; and the Nano version focuses on efficient device-side task processing. Bloomberg predicts that the official announcement of the related transaction may have to wait until the WWDC conference in June. As for the choice of partners, Apple may consider OpenAI or other providers.