BabyAGI
The viral 2023 agent loop, reborn as a self-building framework
Overview
BabyAGI began as Yohei Nakajima's viral 2023 task-loop agent script. That code is archived; the repo now hosts 'functionz', an experimental framework for self-building agents that store and run functions from a database. Educational by design and explicitly not production-ready.
Pricing
Pricing shown for reference only. These figures reflect RECATOOLS research as of 11 Jul 2026 and may be out of date or incomplete. This is not financial or purchasing advice — always confirm the current price on the provider’s official website before making any decision.
Use cases
ASEAN Perspective
BabyAGI in Southeast Asia
ASEAN-region availability and pricing notes coming soon. Drop the editorial team a note via /contact/ if you can supply local context (Singapore/Malaysia/Indonesia/Thailand/Vietnam).
BabyAGI earned its place in history — the March 2023 task-creation, prioritisation and execution loop shaped how a generation of agent frameworks think, and echoes of it survive in CrewAI and LangGraph. But the famous script no longer lives here: it was archived to a separate babyagi_archive repo in September 2024, and the main repo now hosts 'functionz', Nakajima's experimental framework for self-building agents that store and execute functions from a database, with a dependency graph and dashboard. He is blunt that it is not meant for production. Treat the score accordingly: outstanding as a teaching artefact, still updated (commits into January 2026), but nobody should ship it. Teams needing a maintained agent platform belong on LangGraph, CrewAI or a commercial stack. It runs locally with your own keys, so ASEAN access is a non-issue.
What people say
Nobody should confuse a landmark with a product, and BabyAGI is the clearest example in this directory. Yohei Nakajima's March 2023 script — an LLM loop that creates tasks, executes them, spawns new ones from the results and re-prioritises the queue — went viral alongside AutoGPT and shaped how CrewAI, LangGraph and dozens of successors think about agents. That's the landmark part.
Here's the catch for anyone arriving via the GitHub link: the famous script isn't there anymore. It was archived to a separate babyagi_archive repo as a September 2024 snapshot, and the main repository now hosts something different — 'functionz', an experimental framework for self-building agents that stores functions in a database, tracks their imports and dependencies as a graph, and ships a dashboard plus a chat playground. There's also babyagi-2o, a sibling repo chasing the smallest possible self-building agent. The project still draws attention (22.3k stars) and Nakajima was committing as recently as January 2026, so it isn't dormant — but he stays disarmingly honest about what it is, warning it's 'not meant for production use'.
Run it and you'll learn a lot about why agents fail: loops that burn API credits without converging, tasks that drift from the objective, context that evaporates between steps. That's genuinely the value — it's a teaching skeleton small enough to read in one sitting. Prototype on it, study it, steal its ideas. Then build the real thing on a maintained stack.
Summary of public user & expert reviews, compiled by RECATOOLS.
Notable facts
- BabyAGI was created by a venture capitalist (Yohei Nakajima) as a weekend experiment to test whether LLMs could autonomously pursue goals — it became one of the most-starred AI repositories in GitHub history.
- The entire BabyAGI codebase is under 200 lines of Python — an extraordinarily simple implementation for a concept that influenced much larger frameworks.
- BabyAGI's academic paper on task-driven autonomous agents has been cited in foundational research on LLM agent architectures.
Frequently asked questions
About this listing
This entry was compiled from publicly available data including BabyAGI's official website, press releases, documentation, and reputable third-party publications. RECATOOLS is not affiliated with BabyAGI unless explicitly stated.
Third-party AI tools update their pricing, features, availability, and policies frequently. Information here may be outdated by the time you read this — we make reasonable efforts to keep listings current, but cannot guarantee absolute accuracy.
For the latest details, please refer to BabyAGI directly →
Spotted something out of date? Suggest an update →
BabyAGI in the news
Alternatives to BabyAGI
More in Agents & Automation
