docker model install-runner

Description Install Docker Model Runner (Docker Engine only)
Usage docker model install-runner

Description

This command runs implicitly when a docker model command is executed. You can run this command explicitly to add a new configuration.

Options

Option Default Description
--backend Specify backend (llama.cpp|vllm). Default: llama.cpp
--debug Enable debug logging
--do-not-track Do not track models usage in Docker Model Runner
--gpu auto Specify GPU support (none|auto|cuda|rocm|musa|cann)
--host 127.0.0.1 Host address to bind Docker Model Runner
--port Docker container port for Docker Model Runner (default: 12434 for Docker Engine, 12435 for Cloud mode)