Agent-One-Lab/AgentFly
Scalable and extensible reinforcement learning for LM agents.
AgentFly helps AI researchers and developers train large language model (LLM) agents to perform complex, multi-step tasks. You provide a dataset and define specific tools and reward functions, and the system outputs a trained agent capable of using these tools to achieve goals, even with visual or multi-turn interactions. This is ideal for those building sophisticated AI assistants or automated systems.
111 stars.
Use this if you are an AI researcher or developer looking to enhance LLM agents with reinforcement learning, allowing them to make better decisions, use external tools, and handle multi-turn conversations or multimodal inputs.
Not ideal if you are looking for a pre-trained, ready-to-deploy LLM agent solution or if you lack experience in machine learning and agent development.
Stars
111
Forks
10
Language
Python
License
Apache-2.0
Category
Last pushed
Mar 12, 2026
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/agents/Agent-One-Lab/AgentFly"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
lealone/Lealone
一个安全的能够自我进化的 AI 应用开发平台
business-science/ai-data-science-team
An AI-powered data science team of agents to help you perform common data science tasks 10X faster.
langchain-ai/open-swe
An Open-Source Asynchronous Coding Agent
microsoft/UFO
UFO³: Weaving the Digital Agent Galaxy
camel-ai/oasis
🏝️ OASIS: Open Agent Social Interaction Simulations with One Million Agents.