# metauto-ai/GPTSwarm

🐝  The First Self-Improving agents with RL / Prompting Optimization

Repository: https://github.com/metauto-ai/GPTSwarm
Canonical: https://ross.abutalabs.com/products/gptswarm
Homepage: https://gptswarm.org
Language: Python
License: MIT
License Family: permissive
Topics: agent, ai, gpt, python, reinforcement-learning, self-improvement, swarm-intelligence, multi-agent, society-of-mind
Last push: 2026-02-05T10:39:09+00:00

## Health v2 (maintenance only)
Score: 55/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 66, release rhythm 35, longevity 65
- inputs: {"age_days": 919, "days_push": 209, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1040, forks 106 (observed 2026-08-28T04:03:20.371114+00:00)

## What it is
GPTSwarm is a Python library for building LLM-based agents as computational graphs, with modules for agent graphs, memory, LLM backends, and optimizers. It supports automatic self-organization of agent swarms with self-improvement via reinforcement learning and prompt optimization.

## Use cases
- build LLM agents as graphs
- optimize agent workflows with reinforcement learning
- run multi-agent swarms that self-improve
- automatically optimize prompts for LLM agents
- research swarm intelligence with language models
- track LLM operational costs across agent runs

## When to choose
- you want graph-structured LLM agents with automatic optimization
- you need multi-agent self-improvement backed by published research (ICML 2024)
- you want to experiment with swarm intelligence over LLMs

## When to avoid
- you need a simple production chatbot framework without optimization overhead
- you want a framework with broad ecosystem integrations and large community
- your project avoids research-oriented libraries with evolving APIs

## Facets
- artifact type: library
- maturity: active
- function: agent-framework, llm-inference, machine-learning, reinforcement-learning, prompt-engineering, rag
- domain: artificial-intelligence, large-language-models, reinforcement-learning, machine-learning
- platform: python
- tags: multi-agent, swarm-intelligence, graph-based-agents, self-improvement, society-of-mind, llm-optimization, research, ai-agents

## Member repositories
- metauto-ai/GPTSwarm (main) score 55

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:03:20.371114+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T07:03:30.735578+00:00, confidence not recorded.
  - readme: https://github.com/metauto-ai/GPTSwarm (fetched 2026-08-28T04:03:20.371114+00:00, sha 511d3ac6b9ae)
- Data as of 2026-08-30T08:39:29.467469+00:00.