Simple-Efficient/RL-Factory
A reinforcement learning framework for training AI agent models with tool-calling capabilities.
Not currently ranked — collecting fresh signals.
star history
RLFactory is an RL post-training framework designed for agentic learning. It decouples the environment from training, allowing users to define tool-use environments and reward functions for agent training. The framework supports async tool-calling for 2x faster training, multi-turn agent interactions, and model judge rewards. It natively supports DeepSearch training and integrates with MCP tools for Qwen3-based agent development.
Frequently asked
- What is Simple-Efficient/RL-Factory?
- A reinforcement learning framework for training AI agent models with tool-calling capabilities.
- Is RL-Factory open source?
- Yes — Simple-Efficient/RL-Factory is open source, released under the Apache-2.0 license.
- What language is RL-Factory written in?
- Simple-Efficient/RL-Factory is primarily written in Python.
- How popular is RL-Factory?
- Simple-Efficient/RL-Factory has 1.8k stars on GitHub.
- Where can I find RL-Factory?
- Simple-Efficient/RL-Factory is on GitHub at https://github.com/Simple-Efficient/RL-Factory.