BabyAGI is an experimental open-source framework that uses LLMs to autonomously break down high-level goals into subtasks, prioritize them, execute them, and adapt based on results. Originally released in 2023 by Yohei Nakajima, it pioneered the task-driven autonomous agent pattern. It is primarily a developer research tool and not intended for production use. The system stores task history and function outputs in a database for iterative self-improvement. Key features: - Autonomous task creation and prioritization loop - Stores and retrieves function/task results from a database - Dashboard to manage, update, and view function activity - Integrates with external APIs and tools for task execution - Open-source with no licensing cost (users pay LLM API costs)
Free and open-source. Users pay only for LLM API usage (e.g. OpenAI).