Integradio Handbook

Welcome to the Integradio Handbook — your complete guide to building Gradio applications with semantic intelligence.

Integradio extends Gradio with vector embeddings powered by Ollama and nomic-embed-text. Components carry semantic intents, making them discoverable by meaning rather than by ID or label alone.

What you’ll learn

• Getting Started — Install Integradio, set up Ollama, and wrap your first component
• Semantic Wrappers — The core semantic() function and all specialized wrappers for complex components
• Page Templates — 10 pre-built layouts for common UI patterns (Chat, Dashboard, Gallery, and more)
• Visualization — Mermaid diagrams, D3.js interactive graphs, ASCII art, and FastAPI integration
• Reference — SemanticBlocks API, architecture internals, events, WebSocket mesh, and security scope

Who is this for?

• Python developers building Gradio UIs who want AI-discoverable components
• ML engineers who need programmatic access to component graphs for agent workflows
• Teams looking for pre-built page templates with semantic search baked in

Quick taste

from integradio import SemanticBlocks, semantic
import gradio as gr

with SemanticBlocks() as demo:
    query = semantic(
        gr.Textbox(label="Search"),
        intent="user enters search terms"
    )
    demo.search("user input")  # Finds it by meaning
    demo.launch()

Everything runs locally. No cloud APIs. No tokens. No latency.