Latest News
Contracts · Milestones · Announcements
Ranvier to Launch MODELIST
TOPEKA, Kan. — Ranvier, LLC, a data analytics and machine learning software company, will debut MODELIST at the 2026 AI+ Expo (May 7–9, Walter E. Washington Convention Center) ahead of its commercial launch this summer. MODELIST is a no-code predictive AI platform that lets organizations build, validate, tune, and deploy custom machine learning models in minutes through a graphical user interface (GUI) — without writing a single line of code, and entirely within their own environment, with no external cloud dependency and no ML engineering required.
MODELIST was developed and validated under a completed U.S. Air Force Direct-to-Phase-II SBIR contract (Topic AF251-D016) with the Air Force Nuclear Weapons Center. Ranvier delivered an ensemble neural network predictive reliability model for the B61-12 Tail Kit Assembly (TKA) — integrating real-world and counterfactual components to produce near-real-time reliability forecasts — and achieved over 98% predictive accuracy, replacing legacy reliability techniques. The same platform powering that mission-critical defense work is what Ranvier is bringing to market for government, healthcare, financial services, and enterprise customers.
The platform is designed around four principles that traditional ML tools and AutoML platforms force buyers to trade off:
Speed. Users move from raw data to deployed model in minutes, not weeks — without writing code or maintaining costly ML engineering resources.
Accuracy. MODELIST is built for iteration. Users can develop, test, deploy, and re-tune models in a continuous in-platform workflow — pushing each model toward its performance ceiling instead of accepting an AutoML default. Tuning cycles that take weeks on traditional platforms happen in minutes inside MODELIST, which is how Ranvier customers reach mission-grade accuracy on problems where one-shot models fall short.
Control. Customers fully own and shape how their models behave. No black boxes, no vendor lock-in, no opaque automation.
Security. MODELIST runs entirely within the customer’s environment. Sensitive data never leaves the premises, making it suitable for regulated, classified, and air-gapped use cases.
MODELIST’s performance extends beyond defense. In a clinical proof-of-concept, the platform achieved 92% accuracy in Alzheimer’s diagnosis using minimal patient data — work recognized with the Best New Investigator Poster Award at ISPOR 2024.
“We built MODELIST because the organizations that need predictive AI the most — defense, healthcare, finance — are the same ones locked out of it by cost, complexity, and security constraints,” said Cade Graber, Founder and CEO of Ranvier. “Our customers shouldn’t have to choose between speed, accuracy, control, and data security. MODELIST gives them all four — and the in-platform tuning workflow to keep pushing model performance higher long after deployment.”
Ranvier will be available throughout the AI+ Expo to demo MODELIST and discuss applications in reliability prediction, mission planning, logistics forecasting, clinical outcomes, fraud detection, predictive maintenance, and more.
About Ranvier
Ranvier, LLC is a Topeka, Kansas-based software company building MODELIST, a no-code predictive AI platform that lets users build, tune, and deploy machine learning models without writing a single line of code — built for organizations that require speed, accuracy, transparency, and full data control. Ranvier recently completed a U.S. Air Force SBIR Phase II delivering an ensemble neural network reliability model with >98% accuracy for the B61-12 program and is preparing MODELIST for commercial release in summer 2026. Learn more at ranviersoftware.com.
Ranvier Selected by U.S. Air Force for Phase II SBIR Contract
Topic Number: AF251-D016
Ranvier, LLC has been selected to develop the Twin Neural Network Reliability Solution for predictive modeling of the B61-12 Tail Kit Assembly, leveraging its proprietary software platform, MODELIST.
This initiative builds upon a successful Alzheimer’s Dementia proof-of-concept study, in which MODELIST enabled the creation of a bespoke neural network model that achieved 92% accuracy in diagnosing AD using minimal patient data — demonstrating MODELIST’s ability to rapidly create high-performing machine learning models.
The primary objective is to develop a twin neural network model for predictive reliability modeling, replacing outdated techniques to better assess and predict the reliability of the B61-12 Tail Kit Assembly. The model integrates real-world and counterfactual components for near-real-time reliability predictions.
A secondary objective focuses on refining MODELIST for commercialization, with enhancements to the user interface and experience to ensure it is intuitive, user-friendly, and ready for widespread adoption across industries.
At the conclusion of the project, Ranvier will deliver a fully developed MODELIST platform to the USAF. With global demand for predictive analytics projected to grow at a CAGR of 28.3% from 2025 to 2030, MODELIST’s combination of accessibility, power, and affordability positions it as a transformative solution in the predictive analytics market.
Have a question?
Let's talk.
We’re always available to discuss our products or answer questions.