CLI reference

Every vllm-helper command, flag by flag. Generated from the CLI itself.

Usage

vllm-helper

Description

vllm-helper -- serve large models with vLLM on one GPU machine or
several (models, switching, sharding, replica pools, recipes, optional
Coder Agents sync). Per machine, only one native model is loaded at a time.

Subcommands

Name	Purpose
`adopt`	Detect an already-running vLLM server and save it as a managed profile
`coder-sync`	Manually sync a profile's Coder model entry
`doctor`	Diagnose hardware, running servers, integrations, and config drift
`edit`	Open a profile in $EDITOR and validate the result
`inspect`	Show a model's license, weight size against this machine, and detected flags
`install-gguf`	Install the plugin that lets vLLM load GGUF models
`list`	List saved model profiles
`logs`	Show the server log, following by default on a terminal
`new`	Pull a new HuggingFace model, save it as a profile, and switch to it
`node`	Register machines by HTTP URL and SSH target
`pool`	Run a router in front of replicas serving the same model
`recipes`	Tested launch configurations for known-good models
`remove`	Delete a profile, and its Coder model entry when sync is configured
`restart`	Re-apply the active profile and restart the server
`search`	Search HuggingFace and pick a model to pull and switch to
`serve`	Run a web panel for this machine
`set`	Change profile settings from the command line
`setup`	First-time setup: detect hardware and vLLM, write config, offer to install the service
`shard`	Split one model across machines (Ray or vLLM multiprocessing)
`spark`	DGX Spark helpers
`status`	Server state, active model, and whether the endpoint responds
`stop`	Stop the server, keeping profiles and config
`switch`	Switch to a saved profile, restarting the server
`upgrade`	Update vllm-helper itself from a published release
`version`	Print the version