How to keep an OpenAI client working with no model

POST a standard chat body. You get a well-formed chat.completion with zeroed usage and a fixed content string.

Run it

curl -s -X POST https://api.model.tools/v1/chat/completions -H 'Content-Type: application/json' -d '{"model":"any","messages":[{"role":"user","content":"hi"}]}'

Response

{"object":"chat.completion","usage":{"total_tokens":0}}