Ant Group’s AQ Medical AI Platform Wins Top 30 'SAIL Award' at 2025 World Artificial Intelligence Conference}
Ant Group’s autonomous 'AI-driven full-process health management platform' AQ was named among the top 30 at the 2025 WAIC 'SAIL Award,' showcasing its impact on AI healthcare services for over 800 million users.
Recently, the 2025 World Artificial Intelligence Conference (WAIC) announced the Top 30 list for the highest award, the SAIL Award. Ant Group’s self-developed "AI-driven full-process health management platform" was selected, alongside winners like Xiaomi and iFlytek. Currently, this platform is applied in Ant Group’s AI health application AQ and its partners, providing comprehensive AI medical services for over 800 million users and thousands of healthcare institutions.
The SAIL Award (Super AI Leader Award), established in 2018, is awarded by authoritative experts including academicians from the Chinese Academy of Sciences and the Chinese Academy of Engineering. It recognizes projects that make groundbreaking technological and application innovations in AI that significantly impact human life.
The "AI-driven full-process health management platform" relies on Ant Group’s proprietary medical large model, built on three core technologies: medical science Q&A, disease diagnosis and treatment, multimodal real-time interaction, forming a comprehensive service system connecting institutions, doctors, and users.
As the technical backbone, Ant’s medical large model has recently upgraded to a new generation of multimodal medical reasoning capabilities. On June 24, the authoritative MedBench platform updated its evaluation results, showing that Ant’s medical large model scored 98.9 and 94.7 points, winning both self-test and evaluation benchmarks. In the OpenAI-led HealthBench assessment, its complex reasoning ability (HealthBench Hard score 0.2702) and medical consensus ability (HealthBench Consensus score 0.9275) surpassed mainstream models like DeepSeek.
Recently, at KDD 2025, the top international AI conference, Ant’s team introduced a novel multi-image document understanding method, ADMIRE architecture, which was selected for publication. Tests on four open datasets with two mainstream models set new industry benchmarks in medical visual document analysis.
Currently, this technology has been integrated into AQ, greatly improving its report interpretation accuracy and reasoning efficiency. It also offers a new plug-and-play multimodal recognition solution, addressing industry pain points such as image redundancy, low reasoning efficiency, and poor results in multi-image rich text scenarios.
Ant Group’s health division has been established for 11 years. Since entering the AI healthcare field in 2023, it has been committed to building interpretable, trustworthy, and practical large models. On June 26, it officially launched the standalone AQ app, promoting AI technology to serve every family’s health management.