Expand description
Managed-mode launch plan: argv from the spec and the resolved model, environment, working directory, and the “is something already on this port” guard that keeps the adapter from ever starting over a server it does not own.
Structs§
- Launch
- A launch ready to spawn.
Functions§
- build_
launch - Build the launch for a llama.cpp or mlx-lm instance. The argument shape
follows the model lab’s launchers (
serve-llamacpp <model.gguf> [--draft m] [--kv q] [--ctx n] [--port p] [--host h],serve <model> --port p --host h); it lives in this one function. - check_
format - The model must be in a format the runtime loads.