exo dedicated-inference model create

Description

This command creates an AI model by downloading it from Huggingface.

The name parameter must be a valid Huggingface model name (e.g. mistralai/Mixtral-8x7B-Instruct-v0.1).

exo dedicated-inference model create NAME (e.g. swiss-ai/Apertus-8B-Instruct-2509) [flags]

Options

OptionDescription
--help, -hhelp for create
--huggingface-tokenHuggingface token if required by the model
--zone, -zzone

Options inherited from parent commands

OptionDescription
--config, -CSpecify an alternate config file [env EXOSCALE_CONFIG]
--output-format, -OOutput format (table|json|text), see "exo output –help" for more information
--output-templateTemplate to use if output format is "text"
--quiet, -QQuiet mode (disable non-essential command output)
--use-account, -AAccount to use in config file [env EXOSCALE_ACCOUNT]

Related Commands

Last updated on