Menu
TBP Community
Our Open-Source Research Initiative
The framework of the Thousand Brains Project is a foundational technology for the future of artificial intelligence based on the neocortex. Because Numenta is committed to making this technology accessible to everyone, all of the project’s software and ongoing research is open source under the MIT license. This allows you to work with our technology in whatever way works best for you ‒ learn about the theory, dive into the source code, or start your own implementation.
The first implementation of this project “Monty”, in reference to Vernon Mountcastle, who proposed the columnar organization of the mammalian neocortex, is one specific instantiation of the Thousand Brains Theory. In the future, there may be other implementations as part of this project.
The ultimate aim is to enable developers to build AI applications that are more intelligent, more flexible, and more capable than those built using traditional deep learning methods. Monty is our first step towards this goal and represents an open-source, sensorimotor learning framework.
About the Community
Our community is an eclectic collection of researchers, developers, scientists and hobbyists interested in building biologically-inspired intelligent systems based on the Thousand Brains Theory. To get involved, join our Discourse here.
We are also looking for closer collaboration opportunities with other research labs or application teams. If you are interested, please send us an email at ThousandBrains@numenta.com.
Join Our Discourse Forum to Get Involved
This is the best place to ask questions, search for answers, or just interact with others working on similar problems. We aim to be a welcoming community and hope you’ll join us!
Meetups, Hackathons, & Other Live Events
We hold community events like meetups, hackathons, and workshops throughout the year. All our live events are scheduled via Meetup, which are free and open to anyone.
Get Started with Our Documentation
If you’re new to the project, a great place to start is by reading our documentation, which covers every aspect of this project in detail, including the project’s vision and core ideas.
Subscribe for Thousand Brains Project updates
Why Contribute?
Download this two-pager to learn more about the Thousand Brains Project and why you should contribute.
Get Started on our YouTube Channel
Explore our YouTube content, which is categorized into core concepts, brainstorming sessions, and paper reviews. To jumpstart your journey, check out our curated Quick Start Playlist here:
Getting Started
7 Videos
Viviane Clay, the director of the Thousand Brains Project, introduces you to the YouTube channel and its playlists and explains what this open-source project is all about.
This video covers the inception of the Thousand Brains Project, emphasizing a new ‘cortical messaging protocol’ (called ‘AI bus’ in this video) and the difference between structured and unstructured AI models. It explores cortical columns, sensorimotor learning, and the architecture of the brain.
Viviane Clay covers the Cortical Messaging Protocol (CMP, referred to here as Common/Cortical Communication Protocol), the critical idea that enables all modules in the Monty system to communicate. Developed from early ideas about a shared “AI Bus,” the CMP standardizes interactions between Monty’s sensors, learning modules, and motor systems, making Monty flexible and modular.
Viviane presents a comprehensive overview of how Monty currently works and goes into depth on how the EvidenceLM works in particular. She also explains how voting is implemented using the EvidenceLM.
The team runs the first live demo of Monty on real-world data! It’s a very special moment. They show that models trained in simulation can be recognized in the real world despite lighting conditions and all sorts of noise. The demo was developed during a one-week hackathon.
In this presentation, Viviane takes us on a deep dive into Monty and its sensorimotor modeling system. It provides an overview of its current capabilities and limitations, showcasing the advancements made over three years. We cover object and pose detection, modular structure, learning efficiency, multi-object environments, and real-world sensor integration.