Using SkyRL as a Library

To use SkyRL-Train as a dependency in your project, configure your pyproject.toml with uv. You can install from either PyPI (stable releases) or Git (latest development version).

Some additional dependencies that depend on custom indexes/wheels, such as flash-attn and flashinfer-jit-cache. Below is an example configuration using uv:

[project]
dependencies = [
    # Option A: Install from PyPI (stable release)
    "skyrl-train[vllm]==0.3.1",
    # Option B: Install from Git (latest)
    # "skyrl-train[vllm]",

    "flashinfer-jit-cache==0.5.3",
    "flash-attn==2.8.3"
]

[tool.uv.extra-build-dependencies]
flash-attn = [{requirement = "torch", match-runtime = true}]

[tool.uv.sources]
flashinfer-jit-cache = { index = "flashinfer-cu128" }
# Only needed for Option B (Git install):
# skyrl-gym = { git = "https://github.com/NovaSky-AI/SkyRL", subdirectory = "skyrl-gym" }
# skyrl-train = { git = "https://github.com/NovaSky-AI/SkyRL", subdirectory = "skyrl-train" }

[[tool.uv.index]]
name = "flashinfer-cu128"
url = "https://flashinfer.ai/whl/cu128"
explicit = true

Option A (PyPI): Use the pinned version skyrl-train[vllm]==0.3.1. No additional sources needed.

Option B (Git): Remove the version pin and uncomment the skyrl-gym and skyrl-train sources to install from the latest commit.