Alibaba's new healthcare AI is as good as experienced doctors
What's the story
Alibaba Group Holding has revealed that its healthcare-focused artificial intelligence (AI) model has shown capabilities on par with experienced doctors.
The model, powered by the advanced Qwen series, is now part of Quark, Alibaba's flagship consumer-facing AI assistant app.
The AI system passed China's medical qualification exams, achieving the "Deputy Chief Physician" standard across 12 common medical disciplines.
These include general medicine and surgery, internal medicine, obstetrics and gynaecology, and paediatric medicine.
Performance metrics
Performance and benchmark scores
The healthcare model, which is based on Alibaba's Qwen 2.5-32B foundation model, was trained on extensive, high-quality data and advanced multi-stage training.
According to benchmark data by Quark, the model outperformed DeepSeek's R1 and V3 and OpenAI's GPT-4o in test accuracy.
It scored an impressive 74.8% accuracy at the "Deputy Chief Physician" level and a commendable 56.4% at the top-tier "Chief Physician" standard.
China's medical certification ranks healthcare workers in five levels, placing "Deputy Chief Physician" as the fourth highest.
Strategic partnerships
Deployment and collaboration with healthcare institutions
Alibaba has fully integrated the healthcare model into Quark, which automatically deploys it when users ask health-related questions.
The company has also collaborated with hospitals and medical institutions to refine the model for better accuracy.
These institutions are now adopting this advanced AI system in their own applications, further enhancing its capabilities and reach in the healthcare sector.