vllm-helper new
Pull a new HuggingFace model, save it as a profile, and switch to it
Usage
vllm-helper new <org/model-repo> [flags]
Options
--no-follow
| Type | bool |
restart and return immediately instead of watching the boot
--quant
| Type | string |
which quantization to serve, for repos that ship several (see 'vllm-helper inspect')
--timeout
| Type | duration |
give up waiting after this long (default: wait forever)
--wait
| Type | bool |
block until the endpoint is healthy, without streaming logs