AgileRL/AgileRL

Streamlining reinforcement learning with RLOps. State-of-the-art RL algorithms and tools, with 10x faster training through evolutionary hyperparameter optimization.

63
/ 100
Established

Implements evolutionary hyperparameter optimization that automatically tunes agent networks during training through population-based mutation and selection, eliminating the need for separate HPO runs. Supports diverse RL paradigms—on-policy (PPO), off-policy (TD3, DQN), offline RL, multi-agent (MADDPG, MATD3), and contextual bandits—with distributed training across multiple workers and Petting Zoo compatibility for multi-agent environments. Also includes LLM fine-tuning capabilities via reinforcement feedback (RFT) with optional dependencies for transformers, DeepSpeed, and PEFT.

896 stars. Actively maintained with 10 commits in the last 30 days.

No Package No Dependents
Maintenance 20 / 25
Adoption 10 / 25
Maturity 16 / 25
Community 17 / 25

How are scores calculated?

Stars

896

Forks

66

Language

Python

License

Apache-2.0

Last pushed

Mar 09, 2026

Commits (30d)

10

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/agents/AgileRL/AgileRL"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.