ModelsLive builds grounded multimodal AI agents that understand people, context, and real-world signals across consultation, commerce, and personality support.
About
ModelsLive is built on a simple belief: AI models should operate in the real world, not just in benchmarks. We build grounded multimodal agents that connect language, visual signals, user context, and domain knowledge to create genuine value in everyday human situations.
Our work spans AI consultation, live commerce, and personality support, three domains where the gap between what AI can do in a lab and what it actually delivers to people remains wide. We are closing that gap with deployable, research-backed products.
Products
Each product is grounded in a distinct real-world context, where AI models must read people, context, and domain knowledge to create genuine value.
AI models retrieve domain knowledge, verify evidence, and deliver grounded recommendations users can trust.
AI models read emotional context, adapt to personality and communication style, and sustain relational understanding across extended interactions.
Professional decisions stay with experts, while social intelligence is carried by a digital counterpart. An AI Copilot for high-stakes decisions, starting with medical aesthetics.
Founder
Yuyan holds a PhD in Computer Science from Fudan University and was a postdoctoral fellow at Cornell, working on large language models. She has published 35 papers at top AI venues, 24 as first author, and is the first inventor on 39 granted patents; her research has been cited by Tsinghua, Peking University, Stanford, Harvard and Princeton, and referenced in research at Baidu and Alibaba. She founded ModelsLive to take that work out of papers and into products anyone can use.
Personal homepage →Contact
For partnerships, product trials, or general questions, reach out using the addresses below.
