Introducing RejuveAge-Q & LinAge2: Rejuve.AI’s New Tools for Tracking Biological Age

We are excited to announce not one, but two new biological age estimation tools coming to the Rejuve Longevity App: RejuveAge-Q and LinAge2.

These two open models expand our scientific foundation and give Rejuve community members and researchers new ways to estimate biological age, understand health trajectories, and contribute to collaborative longevity research.

RejuveAge-Q: Accessible, Science-Based Biological Age via Questionnaires

RejuveAge-Q is our newly developed questionnaire-based biological age estimator, designed to put longevity insights in everyone’s hands. The “Q” stands for Questionnaire, and it means that even if you don’t have blood tests or wearables, you can still get a personalized read on your biological age. How is this possible? Our data science team crafted an AI model that derives aging indicators from everyday health questions. In essence, RejuveAge-Q looks at your answers about lifestyle and health history and computes a biological age – an estimate of how old your body is acting, as opposed to what the calendar says.

RejuveAge-Q was born from our commitment to accessibility in longevity science. Not everyone has immediate access to lab tests or advanced diagnostics, but everyone should have the opportunity to understand their health trajectory. By leveraging smart algorithms and simple Q&A, RejuveAge-Q extends longevity insights to a much broader audience. This tool was first showcased by Dr. Macsue Jacques at ARDD 2025 (a leading aging research conference) as a biological clock based on questionnaire data, allowing even those without blood tests to gain personalized health feedback. In true Rejuve.AI fashion, it’s a community-centric innovation: the more people use it and provide feedback, the smarter it gets. Ultimately, RejuveAge-Q embodies our DeSci (Decentralized Science) ethos, turning everyday personal data into collective knowledge that benefits each contributor.

RejuveAge-Q is a machine learning model trained on publicly available NHANES [1] data to estimate biological age using health and lifestyle questionnaires. It doesn't rely on blood tests or wearables, making it highly accessible. The model was designed with missing data robustness and question selection strategies to prevent overfitting to age-revealing inputs. While the current model uses questionnaire inputs alone, we’re already building new capabilities that will allow it to incorporate predictive biomarker proxies, further enriching the insights it delivers.

This model will be available in the Longevity App as a free tool for all users. It's an important step toward democratizing biological aging feedback while reinforcing our open science approach. The model is also open source for researchers and developers who want to explore or build on it.

LinAge2: Clinical-Grade Estimation with Feature Attribution

Originally developed by researchers at the National University of Singapore’s Yong Loo Lin School of Medicine, LinAge2 [2] is an advanced algorithm that calculates biological age using clinical health data. In simpler terms, it’s a clock that takes in an array of health markers (like blood and clinical measurements) and outputs how fast a person is truly aging on the inside, often revealing insights that chronological age alone can’t capture.

Why did we choose LinAge2? Quite simply, because it works. In peer-reviewed studies, LinAge2 has outperformed many existing aging clocks in gauging long-term health outcomes. For example, the researchers showed that LinAge2 predicts 10- and 20-year mortality risk more accurately than one’s birth age or even some DNA methylation (epigenetic) clocks. It also correlates with functional health: people with a younger LinAge2 age tend to walk faster, think more clearly, and handle daily activities more independently than those with an older LinAge2 age. Importantly, LinAge2 doesn’t just give a single age estimate and leave you guessing why. It pinpoints which health factors are driving your aging. In the original study, LinAge2 could highlight specific issues (for instance, smoking habits or metabolic syndrome) that were pushing an individual’s biological age higher, offering actionable targets to improve longevity. This focus on concrete, personalized feedback is a perfect complement to Rejuve.AI’s philosophy of actionable insights.

We want to take a moment to acknowledge the team behind LinAge2. The model was developed by a team led by Associate Professor Jan Gruber at NUS Medicine, with collaborators including Dr. Sheng Fong and renowned longevity researcher Prof. Brian Kennedy, among others. Their work, published in npj Aging in 2025, represents a major leap in clinical aging clocks. By integrating LinAge2 into our platform, Rejuve.AI is not just adopting a new tool, but building a bridge between academic breakthroughs and everyday wellness technology. Our users will benefit from the years of research and validation behind LinAge2, and the scientists get to see their algorithm make a real-world impact through a broader, decentralized user base. It is a promising demonstration of how DeSci can unite labs and laypersons in the fight against aging.

LinAge2 estimates biological age using 57 lab-based biomarkers and provides per-feature attribution, highlighting which biomarkers are driving a user's predicted age. This transparency is rare among aging clocks and fits naturally within our goal of actionable, user-aligned longevity insights. We also added an imputation feature to handle missing lab values, allowing broader use without complete datasets.

LinAge2 will be available in the Longevity App as part of our Premium tier. This is one of several advanced features we’ll be offering to help support the growth of the platform while ensuring core tools like RejuveAge-Q remain open and accessible.

Why This Strengthens the DeSci Ecosystem

Both models are open source and auditable. By contributing interpretable models backed by published data and rigorous training methods, we help shift longevity science toward reproducibility and open collaboration.

DeSci thrives on tools that can be independently tested, remixed, and applied across borders. Researchers can fork and adapt RejuveAge-Q, compare outputs between survey- and lab-based models, or integrate either into their own decentralized health studies.

With both RejuveAge-Q and LinAge2 in our toolkit, Rejuve.AI is entering a new era of personalized longevity insights. What does this mean for you as a user?

For everyday users, these tools offer two complementary paths to understanding your biological age. RejuveAge-Q gives you a quick, accessible snapshot. All it takes is answering some questions to get a sense of where you stand and what you might improve. It is immediate and non-invasive, ideal for regular check-ins. On the other hand, LinAge2 unlocks a deeper level of insight if you have more detailed health data available. As we roll out LinAge2 on our platform, users who input relevant clinical metrics (for example, blood test results from your doctor’s visit or a partner longevity clinic) will receive a LinAge2 biological age assessment alongside their usual app insights. This means that if you are the kind of longevity enthusiast who tracks blood biomarkers, you will be able to see how those translate into predicted aging trajectories according to one of the most advanced clocks in the field. And if you are not collecting lab data, RejuveAge-Q has you covered in the meantime, and it may even prompt you when it is worth getting a test done. Together, these features ensure that whether you are a casual user or a quantified-self biohacker, Rejuve.AI adapts to the data you have and gives you valuable feedback.

Crucially, both tools uphold the high standards of scientific validity and transparency that the longevity and DeSci community expects. We avoid overstating what these clocks can do. They are guidance tools, not crystal balls. Biological age algorithms are continually improving, and they offer probabilities and risk indications, not absolute certainties about your future. We want our community to be empowered by RejuveAge-Q and LinAge2, but also to understand them in context. If RejuveAge-Q highlights an area that may benefit from attention, take it as a prompt to reflect, explore further, or consider steps that could support long-term health. It is not a diagnosis, but a signal that further data or adjustments might be worth considering. Similarly, LinAge2’s age prediction should spark a conversation between you and your healthcare providers about proactive steps, rather than be seen as a fixed verdict. By communicating these nuances, we aim to avoid hype and build trust. Longevity science is a marathon, and informed, engaged users are our best partners in that process.

From a community perspective, the introduction of RejuveAge-Q and LinAge2 marks an exciting convergence of crowdsourced health data and top-tier research. Every time you use RejuveAge-Q, you are not only learning about yourself, but also contributing anonymized data points to our broader research (with your consent). Those data points help us refine the questionnaire model and explore new patterns. And now, with LinAge2 in the mix, our community-contributed dataset can be benchmarked against a proven clinical standard. We can compare how the questionnaire-based ages stack up against the LinAge2 ages for users who have both, helping to improve both models over time. This is the power of Decentralized Science: people everywhere collaborating through their data to accelerate discovery, under ethical guidelines. All our data collection is under IRB-approved protocols for safety and privacy. We believe this kind of collaboration strengthens and democratizes longevity research, making it more inclusive and effective.

We’ve also updated our Technology page to reflect the latest state of our model architecture and research tools: https://www.rejuve.ai/technology.

Looking Ahead

These releases are part of Rejuve.AI's commitment to building practical tools backed by science. They also reflect our broader model stack roadmap. Future models will incorporate multi-modal data, resilience metrics, and personalized risk estimation, all aligned with our research infrastructure.

For now, we invite researchers, developers, and community members to explore the codebases, test the models, and share feedback:

https://github.com/Rejuve/RejuveAgeQ 

https://github.com/Rejuve/LinAge2-Python 

References

1. Centers for Disease Control and Prevention (CDC). National Health and Nutrition Examination Survey (NHANES). https://www.cdc.gov/nchs/nhanes/index.htm 

2. Fong, S., Chin, T. S., Liu, M., Xu, T., Lai, L. W., Kennedy, B. K., & Gruber, J. (2025). LinAge2: providing actionable insights and benchmarking with epigenetic clocks. npj Aging, 11(29). https://doi.org/10.1038/s41514-025-00221-4