rllm
Democratizing Reinforcement Learning for LLMs
Training agents with reinforcement learning while the same code drives evaluation.
You are on Python below 3.11, or you want inference rather than training.
About rllm
rLLM is an open-source framework for training language agents with reinforcement learning. Bring any harness, run it in any sandbox, and switch training backends with one flag — the same agent code drives both eval and training.
rLLM requires Python >= 3.11. You can install it either directly via pip or build from source.
This installs dependencies for running rllm CLI with the tinker backend (single-machine, Tinker API). For other backends:
rllm is an open-source project written primarily in Python, with 5.8k stars on GitHub. It was last updated in September 2026.
uv pip install "rllm @ git+https://github.com/rllm-org/rllm.git"
