Introduction

Welcome to the official Sapientia documentation.

In an era where artificial intelligence is increasingly centralized on the cloud servers of tech giants, Sapientia offers a different infrastructure approach. This platform shifts the computational burden of Large Language Models (LLM) back to the local environment: the user's device.

Sapientia is a Local-First Agentic AI Orchestration Platform. The application is designed to facilitate the execution, management, and orchestration of intelligent AI agents locally, without transmitting conversation data to external servers.

Why Sapientia?

Vision & Philosophy

The name "Sapientia" is derived from Latin, meaning Wisdom. The development of this platform is grounded in the belief that technology must be built upon a strong ethical foundation and guaranteed data security.

This vision emerged as a response to three primary challenges in the modern AI ecosystem:

  1. Privacy Risks: Potential misuse or leakage of personal data during third-party processing.
  2. Internet Dependency: Operational limitations during offline conditions or unstable connections.
  3. Cost Efficiency: API subscription costs that tend to escalate with usage scale.

Sapientia's core philosophy is Local-First. The goal is to democratize access to Agentic AI by ensuring full ownership and control remain in the hands of the user.

Privacy & Security

Security in Sapientia is the foundation of the system architecture, not merely an add-on feature. A strict Zero-Telemetry standard is applied to ensure data sovereignty.

Cloudless Architecture

Unlike wrapper applications that forward messages to external APIs (such as OpenAI or Anthropic), Sapientia possesses its own inference engine embedded within the application. This provides technical guarantees including:

  • 100% On-Device Processing: The entire AI model computation process occurs within the user's device. No data leaves the local environment.
  • Isolated Storage: Conversation history, vector memory (Vector Store), and configurations are stored using a local database.
  • Data Control: Users retain full control over the entire data lifecycle.

Key Features

Sapientia transforms a computer into an independent AI workstation.

1. Visual Workflow Builder

Sapientia provides a Graph Execution Engine with a visual interface for designing agent behavior.

  • Design agent logic using a node system.
  • Support the creation of complex branching logic.
  • Clear separation of duties between "Planner" (Router) and "Executor" (Agent) functions.

2. Hybrid RAG (Retrieval-augmented generation) System

The memory system is adaptive to information source requirements:

  • Vector RAG: Performs semantic search on personal documents (PDF/DOCX) indexed in the local vector database.
  • Web RAG: Utilizes Google Custom Search integration to retrieve up-to-date information from the internet when necessary, which is then processed locally.

3. System Integration

Sapientia is equipped with Shell Console execution and API Call capabilities to enhance agent capabilities in executing system commands, managing files, or performing external API calls directly through the chat interface.

On this page

Why Sapientia?Vision & PhilosophyPrivacy & SecurityCloudless ArchitectureKey Features1. Visual Workflow Builder2. Hybrid RAG (Retrieval-augmented generation) System3. System Integration