Apple Announces GPT5 Will Be Integrated into iOS 26: iOS 26 Will Integrate ChatGPT5

AIbase基地

Published in AI News · 3 minute read · Aug 11, 2025

Recently, Apple announced a major update, stating that the upcoming iOS 26 system will be officially released next month and will integrate one of the most advanced AI models available - OpenAI's ChatGPT-5 model. This move will significantly enhance the performance of Apple Intelligence, marking another important step for Apple in the field of artificial intelligence.

ChatGPT-5 performs exceptionally well, featuring strong reasoning capabilities and practical coding tools. It also has significant improvements in voice interaction and video perception. Compared to the current GPT-4 model supported by Apple Intelligence, ChatGPT-5 is more powerful and efficient. Apple stated that when Apple Intelligence encounters specific requests it cannot handle, it will use the capabilities of ChatGPT-5 to provide users with more accurate and intelligent responses.

iOS 26, Apple

This system integration also brings a series of highly anticipated new features. For example, Apple Intelligence will add real-time translation functionality in FaceTime and messaging apps, making cross-language communication more convenient and smooth; through the Visual Intelligence feature, the overall system content search experience will be greatly improved, allowing users to find the information they need faster and more accurately.

Notably, users do not need an OpenAI account to access these new features. However, users who have linked their OpenAI accounts will enjoy subscription discounts, which undoubtedly provides users with more choices and benefits.

Surpassing OpenAI! Baichuan Intelligence's Open Source Medical Model Baichuan-M2 Achieves Global Leadership

Baichuan Intelligence released the open-source medical model Baichuan-M2, which scored 60.1 on the HealthBench evaluation, surpassing competitors such as OpenAI and becoming a leading open-source medical model globally. The model uses a lightweight design, supports single-card deployment on RTX4090, reduces costs by 57 times compared to similar solutions, and introduces an optimized version that improves processing speed by 74.9%. It particularly focuses on medical privacy needs and supports on-premise deployment. Research shows that its medical data training not only does not weaken general capabilities but also enhances mathematical and writing skills.

Artificial Intelligence Aids Medical Imaging Analysis and Alleviates the Burden on Radiologists

Australian research institution AEHRC is using Vision-Language Model (VLM) technology to revolutionize medical imaging analysis, focusing on developing an AI system that can automatically generate chest X-ray reports. This technology, trained on vast amounts of medical data, has significantly improved report accuracy and is also trying to integrate emergency records to further optimize it. The research team emphasizes that AI aims to assist rather than replace doctors. They are currently collaborating with hospitals to evaluate the differences between AI-generated reports and manual ones. In the future, they will expand clinical trials to verify the universality of the technology. The key breakthrough lies in alleviating the pressure caused by the shortage of radiologists while ensuring

Qwen Code by Tongyi Qianwen offers 2000 free calls per day, more generous than Gemini

Alibaba Cloud's Tongyi Qianwen launches Qwen Code AI programming agent, offering mainland Chinese users 2000 free calls per day and overseas users 1000. The tool supports terminal command line operations, offering higher free quotas and a simpler installation process (just one command) compared to competitors like Google Gemini CLI. Alibaba Cloud aims to lower the barriers to using AI programming tools by removing token limits, simplifying installation steps, and providing GitHub developer resources, in order to capture market share, especially for this purpose.

Ultra-small TTS model Kitten TTS: Only 15 million parameters

The KittenML team released the open-source lightweight text-to-speech model Kitten TTS, which is only 25MB in size and has 15 million parameters. It supports CPU operation without requiring a GPU. The model provides high-quality multiple voice options, optimizes inference speed for real-time synthesis, and comes with an easy installation guide. Currently in developer preview, it will future release full model weights, mobile SDK, and web version, promoting the popularization and application of speech synthesis technology.

GPT-5 Officially Released, OpenAI's Application for GPT-5 Trademark in China Rejected

OpenAI encountered obstacles in applying for the GPT-5 trademark in China, with multiple trademarks under review after being rejected. A GPT-5 trademark applied for by a company in Shenzhen has already been rejected. Although the trademark registration was blocked, GPT-5, as OpenAI's latest AI model, has shown remarkable performance in areas such as health consultation and programming. This incident highlights the intense competition in the field of AI intellectual property rights, with trademark registration becoming a key step for companies to protect their core technologies.