vllm complete¶

Options¶

--url¶

url of the running OpenAI-Compatible RESTful API server

Default: http://localhost:8000/v1

--model-name¶

The model name used in prompt completion, default to the first model in list models API call.

Default: None

--api-key¶

API key for OpenAI services. If provided, this api key will overwrite the api key obtained through environment variables.

Default: None

-q, --quick¶

Send a single prompt and print the completion output, then exit.

Default: None