install-stack-flagos

Install vLLM and FlagGems inside a GPU container by running an automated deployment skill that handles wheel selection, dependency ordering, and mirror detection for the five-package GPU software stack, followed by per-package validation.

The correct dependency ordering for installing vLLM, FlagTree, FlagGems, FlagCX, and vllm-plugin-FL is enforced automatically during the container deployment, ensuring prerequisites are satisfied before each package is installed and validated.

Yes, you can deploy the GPU software stack across diverse container environments and orchestration contexts, as the installation process detects network mirrors and selects appropriate wheels for the specific environment.

Validate a GPU container installation by running the deployment skill, which performs per-package verification and environment checks, then generates a final health report detailing the installation status and package versions.

Yes, GPU stack installation handles network mirrors by detecting available mirrors during the deployment process and selecting the appropriate wheels to ensure reliable access across diverse container environments.