SymbolicAI

SymbolicAI is an open-source neuro-symbolic framework for Python that combines classical programming with the differentiable, context-aware nature of LLMs. Its core abstraction is the Symbol object, which comes in two modes: syntactic (behaves like a normal Python value — string, list, int) and semantic (wired to the underlying neuro-symbolic engine, so it understands meaning and context). Symbols default to syntactic mode for speed and predictability, since Python's overloaded operators (==, ~, &) would otherwise trigger the LLM engine on every comparison; code opts into semantic behavior explicitly, either at creation or via a `.sem` projection, and dot-notation operations like `.map()` automatically switch a symbol to semantic mode. The framework is built to be modular and extensible — developers can write their own engine, host one locally, or interface with external tools like web search or image generation.