SkyRL is a modular, performant reinforcement learning library for LLMs, designed for real‑world agentic workloads. Its modular design enables users to modify anything - add new environments, easily implement improvements such as asynchronous training, heterogeneous hardware support, and more!
SkyPilot makes RL training with SkyRL easy to run and scale with best cost-efficiency:
- Run on any AI infrastructure, including Kubernetes or clouds
- Zero setup — one command takes care of provisioning, setting up and run the training.
Launch a multi‑node GRPO training job on GSM8K using the cheapest available GPUs using the YAML:
export WANDB_API_KEY="xxx"
sky launch -c skyrl train.yaml --secret WANDB_API_KEYMonitor training progress:
sky logs skyrl
Logs of the training jobs
You can also view the job status in the SkyPilot Dashboard:
sky dashboard
Dashboard showing the status of the training job
If Weights & Biases (W&B) is configured, you can monitor the training run:
- Modular design: plug‑and‑play algorithms, environments, and hardware backends
- Scales from a single GPU to multi‑node clusters via Ray + SkyPilot
- Minimal boilerplate: add new environments quickly (often <100 LoC)