ART

Agent Reinforcement Trainer: train multi-step agents for real-world tasks using GRPO. Give your agents on-the-job training. Reinforcement learning for Qwen3.5, GPT-OSS, Llama, and more!

agent

View on GitHub Website

9.2k Stars

805 Forks

9.2k Watchers

Python Language

apache-2.0 License

100 SrcLog Score

Cost to Build

$3.50M

Market Value

$21.44M

How is this calculated?

Growth over time

3 data points · 2026-04-11 → 2026-04-26

Stars Forks Watchers

💬

How do you feel about this project?

Ask AI about ART

Question copied to clipboard

What is the OpenPipe/ART GitHub project? Description: "Agent Reinforcement Trainer: train multi-step agents for real-world tasks using GRPO. Give your agents on-the-job training. Reinforcement learning for Qwen3.5, GPT-OSS, Llama, and more!". Written in Python. Explain what it does, its main use cases, key features, and who would benefit from using it.

Question is copied to clipboard — paste it after the AI opens.

How to clone ART

Clone via HTTPS

git clone https://github.com/OpenPipe/ART.git

Clone via SSH

[email protected]:OpenPipe/ART.git

Download ZIP

Download master.zip

Found an issue?

Report bugs or request features on the ART issue tracker:

Open GitHub Issues

Similar to ART

netdata huginn pinpoint amon scouter dd-agent egjs merlin PyGame-Learning-Environment inspectIT goappmonitor mario-ai zappr covertutils Recaf deep-trading-agent trezor-agent PKURemote jvm-profiler stackimpact-go sematext-agent-docker opennars docker-inbound-agent xnumon fusioninventory-agent outis pi-web-agent logdna-agent sarl perfmon-agent