> ## Documentation Index
> Fetch the complete documentation index at: https://docs.swai.app/llms.txt
> Use this file to discover all available pages before exploring further.
# Introducing SWAI
> A next-generation orchestration platform for agentic systems
SWAI is our proprietary AI infrastructure — built from scratch to power the [CTZN](https://www.ctzn.live) ecosystem.
It scores. It adapts. It reacts.
SWAI was designed to scale - modular, reactive, and open to interpretation.
It's not just powering systems.
It's rewriting how they're shaped.
## Why it exists
In a space full of hype and static protocols, SWAI introduces something different: a living system.
One that learns from user behaviour, adapts in real-time, and drives the logic behind every outcome.
CTZN isn't just a social experiment.
It's a platform orchestrated by AI, where attention, visibility, and speculation are turned into programmable incentives.
SWAI makes that possible.
## How it works
Behind the scenes, SWAI is a modular framework with four core components:
Decision logic and reasoning that powers intelligent choices based on context and goals
Multi-level memory (short, long, shared) that enables persistent intelligence across interactions
Event detection and signal routing that identifies important events and directs information flow
Action execution and skill-based responses that translate decisions into real-world actions
Each module works independently but feeds into the same loop:
SWAI collects and processes inputs from various sources
The system evaluates and prioritizes information based on relevance and importance
SWAI executes the appropriate response or action based on its decision-making
The system learns from outcomes to improve future performance
## Powered by the best tech
SWAI is designed to leverage the best AI models and technologies available, adapting to different use cases with the optimal provider for each scenario.
Our architecture allows seamless integration with:
* **Foundation models** from leading providers for different reasoning capabilities
* **Specialized models** optimized for specific tasks like vision, audio, and code
* **Open-source alternatives** for flexibility and customization
* **State machines** for reliable, deterministic behavior
This multi-provider approach means SWAI can select the right model for each task - using Claude for nuanced reasoning, GPT for creative tasks, Gemini for multimodal processing, or lightweight open-source models for edge deployment - all within the same system.
By combining these technologies intelligently, SWAI delivers superior performance while optimizing for cost, latency, and capability requirements across diverse applications.
