About AMiner
An AI research assistant built by the experts who know scholarly data best.
AMiner is an AI-powered academic research assistant developed by Z.ai, combining cutting-edge language models with a proprietary knowledge graph developed over a decade. Our pioneering work【ArnetMiner】, received the ACM SIGKDD 2020 Test-of-Time Award — recognizing its long-term impact on academic data mining.
Since its launch in 2006, it has served tens of millions of users across 220 countries and regions, provided push services to more than 400 journals worldwide, and owns all core intellectual property related to the platform.
In September 2025, AMiner underwent a major upgrade, integrating Z.ai's self-developed GLM series of large language models, a repository of over 300 million research publications, and full-process research tools, covering the entire research lifecycle — search, reading, writing, and experimentation. It has since become Z.ai's next-generation AI research assistant, offering academic search, research agents, scholarly resources, and personalized academic spaces under one roof.
What can AMiner do?
AMiner provides a comprehensive suite of AI-powered research tools for researchers at all levels, including literature search, academic Q&A, Deep Research, intelligent document parsing, and document library management.
Frequently asked questions
What types of research data does AMiner cover?
| Data Type | Quantity |
|---|---|
| Paper records | 330 million |
| Scholar profiles | 60 million |
| Patent records | 180 million |
| Citation relationships | 2.8 billion |
| Journals and conferences | 110,000 |
| Knowledge entities | 8.79 million |
| Datasets | Open-source academic databases covering social networks, scholar profiling, cross-source discovery, machine learning knowledge graphs, patents, Dynamic coauthor, and more |
AMiner's research data includes 330 million paper records, 60 million scholar profiles, 2.8 billion citation relationships, 180 million patents, and 8.79 million knowledge entities.Statistics are updated continuously; as of June 2026.
How does AMiner acquire academic data?
- Integrating data sources and indexing papers through data mining and social network analysis and mining technologies.
- Collaborating with scholars and academic institutions to share paper and scholar data.
- Licensing copyright content from content providers.
Is the content generated by AMiner reliable?
The primary research data generated by AMiner is sourced from authoritative public literature repositories such as arXiv and PubMed. Leveraging its proprietary knowledge graph technology, AMiner models and visualizes citation relationships among papers, collaborations among scholars, affiliations between scholars and institutions, and the evolution of disciplines and research topics, building a dynamic knowledge graph of tens of billions of triples covering scholars, journals, conferences, patents, and disciplines — ensuring that content is traceable and evidence-based as much as possible. However, we do not make any commitments or guarantees regarding the validity, accuracy, correctness, reliability, completeness, or timeliness of the page data.
Does AMiner offer enterprise or university integration services?
Stay tuned, or send your requirements to report@aminer.cn.
What model does AMiner use?
AMiner is an academic research platform under Z.ai. Its model capabilities are continuously updated with the latest flagship models in the GLM series, ensuring that the full-featured, full-powered flagship GLM model is embedded promptly, providing users with cutting-edge academic search, analysis, and research support capabilities. The currently used model can be checked and switched via the model selection dropdown in functions such as Academic Search and AI Reading.