Unity-Technologies/ml-agents
Machine learning agents toolkit for Unity
- Chart
- Not on a chart right now
- Stars
- 19.9k
What it is
An open-source toolkit that enables games and simulations to serve as environments for training intelligent agents. It supports training agents in Unity using deep reinforcement learning and imitation learning.
Who it is for
Game developers and researchers building or studying intelligent agents in Unity environments.
Why it is trending
It gained 131 stars today, reaching 19,879 total stars. Started in 2017, its latest release (ML-Agents Release 23) was published in September 2025.
Main features
- Games and simulations as agent training environments
- Deep reinforcement learning and imitation learning support
- Unity and Python integration for training
- Package documentation with tutorials and API references
Summarized on November 17, 2026 from the README and repository data. It can be wrong; check the source before relying on it. Report a wrong summary