FractalBrainOS

FractalBrainOS v5.2 — Kubera Edition

An open-source self-learning neuromorphic engine in C++17. MIT license.

It is an engine, not a product.

Like a combustion engine — it works, but what you build around it is up to you. Connect sensors and motors, and it can learn to control a drone. Connect an LLM, and it can speak. The same code, any application.

How it works

Hierarchical oscillatory neural network using Kuramoto synchronization. Three oscillators form a "triplet" (one neuron). Triplets form higher-level triplets, fractally, to any depth.

Key mechanisms:

• STDP continuously updates connection weights
• Free Energy (Friston) drives unsupervised learning
• Active Inference: predicts its own state and nudges toward it
• Delay-line memory: standing-wave storage, practically eternal
• Clock oscillators: time as memory address
• Homeostasis: prevents epileptic synchronization
• int16 quantization: 75% RAM reduction
• Sleep consolidation: prunes weak connections, strengthens useful ones

What works

The core compiles, runs, synchronizes, learns through STDP, reduces free energy, stores patterns, enters sleep, and consolidates memory. Audited: 29/29 tests passed.

What awaits integration

Some modules are hooks: VSA, full Active Inference, P2P-to-core bridge, LLM integration in the binary. Honest about this — see README.

Kubera Dharma Protocol

Seven ethical principles hardcoded in architecture: Ahimsa, Satya, Dharma, Unity, Seva, Shanti, Svadhyaya.

The engine is built. The rest is up to you.

GitHub: https://github.com/NineNi999neNine/FractalBrainOS