Kaonomi.net: Distributed AI and meta AI
At Kaonomi.net, we believe the next leap in AI will not come from one ever-bigger model, but from many intelligences working together. Two projects bring this vision to life.
M-A-R-I: the engine
M-A-R-I is an open-source framework (MIT licence) for building distributed multi-agent systems. Built on .NET and Microsoft Orleans, it orchestrates virtual agents across a cluster: sequences, triggers, signals, data streams and a shared blackboard. It also brings in Python code or Docker containers. It runs on Linux, Windows and macOS. see https://www.kaonomi.net/p/m-r-i-multi-agent-augmented-reasoning.html
KAONOMI: the interface
KAONOMI is a multi-agent chatbot powered by M-A-R-I. Behind every message, 13 specialised agents collaborate: guardrails, entity analysis, model selection, emotion engine, forecasting… Meanwhile, cognitive agents (Dreamer, Vision, Reflex, Metacognition) keep working in the background. KAONOMI runs on your local LLMs, so your data stays with you, and opens up to frontier AI models. It is also the single interface for every Kaonomi.net solution.
In short: M-A-R-I makes agents work together; KAONOMI lets you talk to them.
see https://www.kaonomi.net/p/kaonomi.html
