SkyRL
Modular reinforcement-learning stack with a training engine, long-horizon agent layer and Gymnasium-style tool-use environments for LLMs.
Tags
Description
Public Apache-2.0 source. The unified skyrl library combines skyrl-train and skyrl-tx and supports a Tinker API backend on hardware you control. skyrl-agent handles multi-turn long-horizon training and evaluation; skyrl-gym supplies math, coding, search and SQL environments. Repository development and quickstart guides cover source setup and model/compute dependencies. The README documents Harbor integration and an October 16, 2026 Beam support recipe. This is software access, not free hosted GPU capacity.
Agent access
Related
- Harbor: Framework from the Terminal-Bench creators for evaluating and optimizing agents and models in sandboxed environments, run in parallel across cloud sandbox providers.
Guides & comparisons
- RL environment libraries compared — NeMo Gym, OpenEnv and Verifiers (Comparison): Compare NeMo Gym, OpenEnv and Verifiers as libraries for writing and running RL environments: authoring model, isolation, documented trainers, install paths, licences and current versions. Distinct from Harbor/AgentEnv/Seahaven and from the agent RL stacks comparison.
- Fine-tuning frameworks compared — Axolotl, LLaMA-Factory and Unsloth (Comparison): Compare Axolotl, LLaMA-Factory and Unsloth as SFT, LoRA/QLoRA and preference-tuning frameworks for tool-use models: documented methods, config style, chat/tool datasets, hardware backends, licences and current versions. Cross-links the agent RL stacks comparison.
- Agent RL stacks compared — verl, SkyRL, Agent Lightning and ART (Comparison): Choose an agent reinforcement-learning stack by the layer you need: distributed training, long-horizon environments, an existing-harness gateway, or a trajectory-and-reward client. Backend ownership matters more than a headline benchmark.
- Post-training a tool-use model — TRL, PEFT, and scale-out RL (Stack): A recommended post-training stack for chat and tool-use models: TRL and PEFT for SFT, LoRA/QLoRA, DPO and GRPO, with OpenRLHF and AReaL as the distributed and asynchronous RL options. Cross-links the fine-tuning frameworks and agent RL stacks comparisons.
Sources
Machine-readable
- JSON
- Markdown
- Source file on GitHub (edit via pull request)