Welcome to Dartantic!

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

Key Features

• Agentic behavior with multi-step tool calling: Let your AI agents autonomously chain tool calls together to solve multi-step problems without human intervention.
• Multiple Providers Out of the Box - OpenAI, OpenAI Responses, Google, Anthropic, Mistral, Cohere, Ollama, and more
• OpenAI-Compatibility - Access to literally thousands of providers via the OpenAI API that nearly every single modern LLM provider implements
• Streaming Output - Real-time response generation
• Typed Outputs and Tool Calling - Uses Dart types and JSON serialization
• Multimedia Input - Process text, images, and files
• Media Generation - Stream images, PDFs, and other artifacts from OpenAI Responses, Google Gemini (Nana Banana), and Anthropic code execution
• Embeddings - Vector generation and semantic search
• Model Reasoning ("Thinking") - Extended reasoning support across OpenAI Responses, Anthropic, and Google
• Provider-Hosted Server-Side Tools - Web search, file search, image generation, and code interpreter via OpenAI Responses, Anthropic, and Google
• MCP Support - Model Context Protocol server integration
• Provider Switching - Switch between AI providers mid-conversation
• Production Ready: Built-in logging, error handling, and retry handling
• Extensible: Easy to add custom providers as well as tools of your own or from your favorite MCP servers

Quick Start

import 'package:dartantic_ai/dartantic_ai.dart';
import 'package:json_schema/json_schema.dart' show JsonSchema;

void main() async {
  // Create an agent with your preferred provider
  final agent = Agent(
    'openai',  // or 'openai-responses', 'google', 'anthropic', etc.
  );

  // Generate text
  final result = await agent.send(
    'Explain quantum computing in simple terms', 
    history: [ChatMessage.system('You are a helpful assistant.')],
  );
  print(result.output);

  // Use typed outputs
  final location = await agent.sendFor<TownAndCountry>(
    'The windy city in the US',
    outputSchema: JsonSchema.create({
      'type': 'object',
      'properties': {
        'town': {'type': 'string'},
        'country': {'type': 'string'},
      },
      'required': ['town', 'country'],
    }),
    outputFromJson: TownAndCountry.fromJson,
  );
  print('${location.output.town}, ${location.output.country}');
}

Documentation

Read the full documentation at docs.dartantic.ai

Contributing & Community

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

Want to chat about Dartantic? 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