GitHub - brekkylab/ailoy: A comprehensive library for building intelligent AI agents and applications

Comprehensive library for building intelligent AI agents

🚀 Quick Start

See how easy to use Ailoy through below examples.

Get your agent just in a single line of code

Check out the simplest python example to build your agent with local models.

import ailoy as ai

# Create an agent with a local model in a single line of code.
agent = ai.Agent(ai.LangModel.new_local_sync("Qwen/Qwen3-8B"))

# Get the response from the agent simply by calling the `run` method.
response = agent.run("Explain quantum computing in one sentence")
print(response.contents[0].text)

Easy to integrate LLM APIs

Here's the simple javascript example with LLM APIs.

import * as ai from "ailoy-node";

async function main() {
  const lm = await ai.LangModel.newStreamAPI(
    "OpenAI", // spec
    "gpt-5", // modelName
    "YOUR_OPENAI_API_KEY" // apiKey
  );
  const agent = new ai.Agent(lm);
  for await (const resp of agent.run("Please give me a short poem about AI")) {
    if (resp.message.contents[0].type === "text") {
      console.log(resp.message.contents[0].text);
    }
  }
}

main().catch((err) => {
  console.error("Error:", err);
});

Browser-Native AI (WebAssembly)

You can build your agent entirely in the browser using WebAssembly just in a few lines of code.

import * as ai from "ailoy-web";

// Check WebGPU support
const { supported } = await ai.isWebGPUSupported();

// Run AI entirely in the browser - no server needed!
const agent = new ai.Agent(await ai.LangModel.newLocal("Qwen/Qwen3-0.6B"));

Quick-customizable Web Agent UI Template

Just Clone to build your own web agent in minutes.

• https://github.com/brekkylab/ailoy-web-ui

🔥 Key Features

Simple Framework and Powerful Features for AI Agents

• No boilerplate, no complex setup
• Reasoning: Extend thinking effortlessly
• Multi-Modal Inputs: Process both text and images
• Extensible Tool Calling: User-defined functions and Model Context Protocol (MCP) tools
• Retrieval-Augmented Generation (RAG): Integrates external knowledge bases without boilerplate

Cross-Platform & Multi-Language APIs

• Provide Python and JavaScript APIs

• Support Windows, Linux, and macOS

• Support Synchronous and Asynchronous APIs

Support Web-browser Native AI (WebAssembly)

• Run AI entirely in the browser - no server needed!

Flexible Model Adoption

• Supports both local AI execution and cloud AI providers
• Effortlessly switch between open-source and AI services
• Minimal software dependencies — deploy anywhere, from cloud to edge

• Fast, memory-safe, minimal dependencies
• Best choice for edge computing and low-resource devices

Documentation & Community

• Documentation (Eng.)

• Documentation (Kor.)

• Discord Community - Join to ask questions, share your projects, and get help.

Example Projects

Installation

Python

Node.js

Browser (WebAssembly)

Support Specifications

Supported AI Models

Type	Provider & Models
Local Model	Qwen3 (0.6B, 1.7B, 4B, 8B, 14B, 32B, 30B-A3B)
Cloud API	OpenAI (GPT)
Cloud API	Anthropic (Claude)
Cloud API	Google (Gemini)
Cloud API	xAI (Grok)

Supported Languags

Supported Platforms