dartantic_ai 0.9.6

A Dart library for easily interacting with LLMs in a typed manner.

Welcome to Dartantic!

The dartantic_ai package provides an agent framework inspired by pydantic-ai and designed to make building client and server-side apps in Dart with generative AI easier and more fun!

Goals

• Agentic behavior with multi-step tool calling: Let your AI agents autonomously chain tool calls together to solve multi-step problems without human intervention.
• Multi-Provider Support: Works with OpenAI, Google Gemini and more
• Type Safety: Leverage Dart's strong typing with automatic JSON schema generation
• Production Ready: Built-in logging, error handling, and provider capabilities detection
• Extensible: Easy to add custom providers and tools

Quick Start

import 'package:dartantic_ai/dartantic_ai.dart';

void main() async {
  // Create an agent with your preferred provider
  final agent = Agent(
    'openai',  // or 'gemini', 'openrouter', etc.
    systemPrompt: 'You are a helpful assistant.',
  );

  // Generate text
  final result = await agent.run('Explain quantum computing in simple terms');
  print(result.output);

  // Use typed outputs
  final location = await agent.runFor<TownAndCountry>(
    'The windy city in the US',
    outputSchema: TownAndCountry.schemaMap.toSchema(),
    outputFromJson: TownAndCountry.fromJson,
  );
  print('${location.output.town}, ${location.output.country}');
}

Key Features

• Streaming Output - Real-time response generation
• Typed Tool Calling - Type-safe function definitions and execution
• Multi-media Input - Process text, images, and files
• Embeddings - Vector generation and semantic search
• MCP Support - Model Context Protocol server integration
• Provider Switching - Switch between AI providers mid-conversation with the full stack of messages, including messages, media and tool calls

Documentation

Read the full documentation

The documentation includes:

• Getting Started Guide - Installation and basic usage
• Core Features - JSON output, typed responses, and DotPrompt support
• Advanced Features - Tool calling, agentic behavior, streaming, and embeddings
• Integration - Logging, model discovery, MCP servers, and custom providers

Installation

Add to your pubspec.yaml:

dependencies:
  dartantic_ai: ^latest_version

Contributing & Community

Welcome contributions! Feature requests, bug reports and PRs are welcome on the dartantic_ai github site.

Want to chat about Dartantic? Then drop by the Discussions forum.

License

This project is licensed under the MIT License - see the LICENSE file for details.

---

Built with ❤️ for the Dart & Flutter community