BEIJING, Oct. 27, 2026 /PRNewswire/ — Wenge AI launched ScienceOra, its research and decision-support platform, on Tencent WorkBuddy, making more than 30 specialist research agents and skills available to users. The initial offering supports literature analysis, hypothesis development, study design, modeling and simulation, validation, and research communication. Users can access ScienceOra through WorkBuddy’s app discovery section for support across the research process, from exploring a new field to conducting advanced research and making informed scientific decisions.
ScienceOra is powered by the ScienceOne Scientific Foundation Model, which was developed as part of a Chinese Academy of Sciences (CAS)-coordinated initiative. Building on that foundation and on Wenge AI’s long-standing work in scientific data, research tools, and decision AI, ScienceOra brings specialized research knowledge, tools, and agent capabilities into one system, extending AI from knowledge retrieval and content generation to the execution of complex research tasks and support for scientific decision-making.
ScienceOra’s initial offering of specialist research agents on WorkBuddy
Covering core scenarios across disciplines
Scientific research involves a great deal of specialized work at every stage, from reviewing the literature and formulating hypotheses to designing experiments, analyzing data, validating findings, and publishing papers. Researchers everywhere face the same questions: how to spot valuable leads in a vast body of literature, how to synthesize scientific evidence and carry complex projects forward, and how to make sound scientific judgments between competing options. ScienceOra on WorkBuddy addresses both general research needs and discipline-specific work, organized around three levels of support — getting started, advanced research, and expert decision-making — for users across disciplines and at different stages of their work.
Getting started in research
At the first level, ScienceOra helps researchers turn a mass of information into a structured understanding of a field. Its agents cover academic literature research, comparative literature analysis, in-depth paper analysis, and literature review mapping; its skills cover literature search, multi-paper comparison, and structured close reading. Together they help researchers trace how a field has developed, compare the core arguments and methods of different studies, and identify research gaps. Beyond finding and summarizing literature, the AI can examine the evidence, conclusions, and limitations of different studies around a specific scientific question, offering reference points for further work.
Advanced research
For advanced research, ScienceOra helps researchers move from analyzing existing knowledge to carrying out specialized tasks. It supports evidence synthesis, study design, and domain-specific data analysis in areas such as systematic reviews and meta-analyses, molecular biology and multi-omics, and drug developability assessment.
In drug development, for example, researchers assessing a candidate molecule’s developability can work with specialist agents to plan experiments, molecular modeling, and simulations, then draw on experimental data to explore validation strategies and opportunities for optimization. ScienceOra also offers manuscript editing, manuscript formatting to meet journal requirements, and research slide deck generation, helping researchers organize, communicate, and present their findings.
Expert decisions
To support expert decision-making, ScienceOra helps researchers weigh scientific evidence and evaluate competing options in rice breeding, materials science, drug development, and research evaluation. Its agents include the Yuandao Siyuan AI breeding scientist, a materials characterization and validation scientist, and an alloy thermodynamics and phase-field simulation scientist.
Yuandao Siyuan is a specialized ScienceOra application for rice breeding. The agent is built on the Yuandao Siyuan intelligent breeding platform, jointly developed by the Hunan Academy of Agricultural Sciences, the Institute of Automation of the Chinese Academy of Sciences, and Wenge AI, and launched this month at the Yuelu Mountain Laboratory in Changsha. The name pays tribute to hybrid-rice pioneer Yuan Longping and evokes the Chinese saying “When drinking water, remember its source.” Drawing on dedicated breeding resources and a knowledge graph, it supports searches for variety approval records and germplasm resources, pedigree tracing, and the tracing of research literature to its sources.
In new materials development, researchers can draw on X-ray diffraction (XRD) phase analysis, structural characterization, and published evidence to compare candidate materials in terms of properties, structural stability, fabrication costs, and potential risks, informing materials selection and subsequent experimental validation.
For research evaluation, a general-purpose reviewer agent and a range of peer-review simulation skills help assess the quality of a study, including its originality, performance against existing benchmarks, methodological rigor, and experimental design. These capabilities help researchers efficiently identify key scientific questions and promising avenues for innovation, supporting the choice of research topics and the evaluation of technical approaches.
ScienceOra’s initial offering of research skills on WorkBuddy
Extending decision AI into research workflows
By combining domain knowledge, scientific data, research tools, and agent capabilities, ScienceOra extends research AI beyond question answering and content generation to complex research tasks and decision support. This brings AI into day-to-day research workflows, helping researchers work more efficiently.
This integration with Tencent WorkBuddy is another step in Wenge AI’s effort to make specialized research AI available to more users. Through the WorkBuddy app ecosystem, ScienceOra brings its agents and skills into everyday workflows, giving researchers, university faculty and students, and other learners more convenient access to research AI tools.
Going forward, Wenge AI will continue to expand ScienceOra’s research capabilities and deepen the integration of decision AI into research settings, making scientific knowledge easier to access, research more efficient, and professional decisions better grounded in evidence.
ScienceOra Website: https://www.scienceone.ai/portal/
Website: https://www.wenge.com/
