Agentic AI: Open-Source and Publicly Available Frameworks and Tools

Open-source frameworks

• AutoGen (Microsoft): Focuses on orchestrating multi-agent systems and enabling conversational workflows between agents. GitHub

• CrewAI: Specializes in creating teams of agents with defined roles and goals to tackle complex tasks. CrewAI

• LangChain: A popular framework for developing LLM applications, allowing for granular control and customization. LangChain

• LangGraph (LangChain Ecosystem): A library for building stateful, multi-agent applications using a graph-based approach. LangChain

• Smol Agents: A lightweight, open-source framework for building AI agents that operate with minimal resources. GitHub 

Platforms and tools with free tiers or trials

• Lindy.ai: A no-code platform for building custom AI assistants (called "Lindies") to automate workflows, with a free tier offering 400 credits per month. Lindy.ai

• Botpress: Allows users to build AI agents with a visual flow editor and offers a free plan with a core builder, 1 bot, and $5 AI credit. Botpress

• Voiceflow: A visual builder for designing conversational AI, offering a free plan including 1 project and a few custom agents. Voiceflow

• AutoGen Studio: Provides a visual interface for creating multi-agent conversations with built-in templates and a free tier with generous usage limits. Microsoft Open Source

• Flowise (Based on LangChain): A visual, no-code tool for creating AI workflows with a free self-hosting option. Flowise

• AgentGPT: Offers a user-friendly interface for building autonomous agents with a free tier providing basic functionality. AgentGPT

• Fin (Intercom): Recognized for its customer service AI agents and offering a free 14-day trial. Fin AI

• Google AI Studio: A platform for experimenting with AI models and building with the Gemini Developer API, offering free usage within specified limits. Google Cloud 

Other noteworthy options

• AutoGPT: An autonomous agent framework that enables self-planning and goal-driven AI agents, with its core codebase being open-source and free to use. GitHub

• Agent2Agent (A2A) Protocol: An open protocol from Google Cloud for facilitating communication and coordination between AI agents across different platforms. World Wide Technology | WWT

• Gemini CLI: An open-source AI agent that brings Gemini's capabilities directly into the developer's terminal, with unmatched usage limits for individual developers. GitHub 

Important considerations:

• LLM API Costs: While the frameworks and platforms themselves might be free or have free tiers, using large language models (LLMs) often involves API calls that may incur costs, especially with higher usage.

• Hosting and Infrastructure: Hosting and deploying AI agents may require infrastructure, according to Budibase.

• Technical Expertise: While many of these tools aim for ease of use, some open-source frameworks may require a certain level of technical knowledge, such as Python proficiency or experience with APIs, says Budibase. 

By exploring these options and carefully considering your needs, technical capabilities, and potential costs, you can find a suitable free or low-cost solution to start building and experimenting with agentic AI.