ramalama-pull - Man Page

pull AI Models from Model registries to local storage

Synopsis

ramalama pull [options] model

Description

Pull specified AI Model into local storage

Options

--authfile=password

path of the authentication file for OCI registries

--help, -h

Print usage message

--tls-verify=true

require HTTPS and verify certificates when contacting OCI registries

--verify=true

verify the model after pull, disable to allow pulling of models with different endianness

Proxy Support

RamaLama supports HTTP, HTTPS, and SOCKS proxies via standard environment variables:

Example proxy URL formats: - HTTP/HTTPS: http://proxy.example.com:8080 or https://proxy.example.com:8443 - SOCKS4: socks4://proxy.example.com:1080 - SOCKS5: socks5://proxy.example.com:1080 or socks5h://proxy.example.com:1080 (DNS through proxy)

SOCKS proxy support requires the PySocks library (pip install PySocks).

See Also

ramalama(1)

History

Aug 2024, Originally compiled by Dan Walsh dwalsh@redhat.com ⟨mailto:dwalsh@redhat.com⟩

Referenced By

ramalama(1).