README.md

---
title: README
emoji: 👁
colorFrom: gray
colorTo: red
sdk: static
pinned: false
short_description: Reactive AI - Reactive Neural Networks and Event-Driven AI
---

<div style="display: flex; gap: 48px;">
  <img src="https://huggingface.co/spaces/ReactiveAI/README/resolve/main/logo-black.png" width=300 />
  <img src="https://huggingface.co/spaces/ReactiveAI/README/resolve/main/logo-white.png" width=300 />
</div>

# Reactive AI
We are working on our own ideas of Reactive Neural Networks (RxNN) and Event-Driven AI, advancing from language models to AGI awareness models.

## Reactive Neural Networks and Event-Driven AI
Reactive Neural Networks (RxNN) are memory-augmented neural networks with higher levels of recurrence (inter-sequence vs. intra-sequence in RNNs),
focused on processing single interactions with access to previous interactions via memory layers. We call this _**event-driven real-time processing**_
to distinguish it from classical _data-driven processing_ of the full conversation history in each interaction. This difference is crucial in case
of AGI and awareness - the key feature of humans awareness, is that we remember what we were doing 10 mins ago, without recalling the whole-day history - we
are working in real-time - just like event-driven _Reactive Neural Networks_.

In Event-Driven AI models are processing the data in reaction to environment or internal events, and are emitting other response events as a result.
Processing of input and output events by the model is called the interaction. Event or an interaction could occur in any point in continous time. Models
have to be stateful and remember the data between the interactions.

_**Strong Reactive Neural Networks**_ like **Reactor** could emit and listen to its internal events, while the _**Weak Reactive Neural Networks**_ are
working only on environment events.


## Stateful Reactive Language Models (RxLM)
Our **Reactive Transformer** and second, improved generation of RxT, are extending stateless Transformer language models (almost all LLMs), introducing
_Attention-based Memory System_ (ABMS) with _Short-Term Memory_ (STM) or multi-level _Mixture-of-Memory_ (MoM / with _Long-Term Memory_). It's based on higher-level
of recurrence and memory - not between tokens like SSMs, Linear Attention (it could be combined with RxLM) or RNNs, but between interactions (query and answer).
They introduce effective stateful processing with continual learning, infinite memory & context and are natively conversational & agentic

## RxLM vs LLM advantages
Processing single interactions in real-time by **Reactive Language Models** leads to **revolutional** improvements in inference speed/cost:
- LLM inference costs are increasing quadratically with conversation length (accumulated for each next message), because of full dialog history processing
- RxLM inference costs are linear, depending only on single interaction tokens (not accumulated) - each next interaction is `number of steps` times cheaper than for LLM
- same for inference speed - LLM has to process full history, while RxLM only single message (only first interaction could be slower because of encoder/memory attention overhead)

> In example, for a dialog with **DeepSeek R1**, that have overally ~90k tokens, I paid for about 1.5M tokens. With **RxLM** it will cost only that ~90k tokens, so it
> will be about **15x cheaper**

## Reactor AGI
Our final goal- **Reactor** - is planned as the first _**awareness AGI model**_, that's modelling consciousness as an _Infinite Chain-of-Thoughts_,
connected to _Mixture-of-Memory (MoM)_ in _Attention-based Memory System_ and _Receptors/Effectors_ systems for real-time reactive processing.
It will be able to constantly and autonomously learn from interactions in _Continouos Live Learning_ process.

[Visit our website!](https://rxai.dev) \[Work in progress\]