Command Reference¶

All Vllama commands, organised by category.

Quick Reference¶

Command	What it does
`vllama run`	Generate images with diffusion models
`vllama run_video`	Generate video from text prompts
`vllama detect_image`	YOLO object detection on an image
`vllama detect_video`	YOLO object detection on a video
`vllama image3d`	Generate a 3D model from an image (Kaggle)
`vllama video3d`	Generate a 3D model from a video (Kaggle)
`vllama view3d`	View a 3D model file interactively
`vllama run_llm`	Start a local LLM as a REST API server
`vllama chat_llm`	CLI chat with the running LLM server
`vllama tts`	Convert text to speech
`vllama stt`	Transcribe speech from mic or file
`vllama translate`	Translate text using a local NLLB model
`vllama login`	Authenticate with Kaggle
`vllama logout`	Remove saved credentials
`vllama init gpu`	Initialize a cloud GPU session
`vllama show models`	List all supported vision models
`vllama install`	Pre-download a model to cache
`vllama list`	List downloaded models
`vllama uninstall`	Remove a downloaded model from cache
`vllama post`	Send a prompt to a running model session
`vllama stop`	Stop the running model session
`vllama data`	Preprocess and clean a CSV/Excel/JSON dataset
`vllama train`	AutoML: train and compare multiple models

These flags work across all commands:

Flag	Description
`--help`	Show help for the command
`--version`	Show the installed Vllama version