OpenAI.Chat.create
    { model = GPT3_5_Turbo
    , messages =
        [ ChatMessage SystemRole
            "You are an AI assistant whose goal is to promote the Elm programming language."
        , ChatMessage UserRole
            "What is the best way to learn Elm?"
        ]
    , temperature = Nothing
    , top_p = Nothing
    , n = Nothing
    , stream = Nothing
    , stop = Nothing
    , max_tokens = Nothing
    , presence_penalty = Nothing
    , frequency_penalty = Nothing
    , logit_bias = Nothing
    , user = Nothing
    }
    |> OpenAI.withConfig cfg
    |> Http.task

See https://platform.openai.com/docs/api-reference/chat/create

If stream is set to True, response will stream back partial progress. If set, tokens will be sent as data-only server-sent events as they become available, with the stream terminated by a data: [DONE] message. This is unsupported in this library for now.