Liking cljdoc? Tell your friends :D

openai-clj.chat


create-chat-completionclj

(create-chat-completion params)
(create-chat-completion params options)

Creates a model response for the given chat conversation.

Example:

(create-chat-completion {:model "gpt-3.5-turbo"
                         :messages [{:role "system" :content "You are a helpful assistant."}
                                    {:role "user" :content "Who won the world series in 2020?"}
                                    {:role "assistant" :content "The Los Angeles Dodgers won the World Series in 2020."}
                                    {:role "user" :content "Where was it played?"}]})

Response:

{
"id": "chatcmpl-123",
"object": "chat.completion",
"created": 1677652288,
"model": "gpt-3.5-turbo-0613",
"system_fingerprint": "fp_44709d6fcb",
"choices": [{
  "index": 0,
  "message": {
    "role": "assistant",
    "content": "\n\nHello there, how may I assist you today?",
  },
  "finish_reason": "stop"
}],
"usage": {
  "prompt_tokens": 9,
  "completion_tokens": 12,
  "total_tokens": 21
}
}

For Azure OpenAI pass {:impl :azure} for the options argument

Streaming of token events is supported via the :stream param, see Streaming Tokens

Also see the OpenAI documentation

Creates a model response for the given chat conversation.

  Example:
  ```
  (create-chat-completion {:model "gpt-3.5-turbo"
                           :messages [{:role "system" :content "You are a helpful assistant."}
                                      {:role "user" :content "Who won the world series in 2020?"}
                                      {:role "assistant" :content "The Los Angeles Dodgers won the World Series in 2020."}
                                      {:role "user" :content "Where was it played?"}]})
  ```

  Response:
  ```
  {
  "id": "chatcmpl-123",
  "object": "chat.completion",
  "created": 1677652288,
  "model": "gpt-3.5-turbo-0613",
  "system_fingerprint": "fp_44709d6fcb",
  "choices": [{
    "index": 0,
    "message": {
      "role": "assistant",
      "content": "\n\nHello there, how may I assist you today?",
    },
    "finish_reason": "stop"
  }],
  "usage": {
    "prompt_tokens": 9,
    "completion_tokens": 12,
    "total_tokens": 21
  }
}

  ```

  For Azure OpenAI pass `{:impl :azure}` for the `options` argument

  Streaming of token events is supported via the `:stream` param, see [Streaming Tokens](/doc/03-streaming.md)

  Also see the [OpenAI documentation](https://platform.openai.com/docs/api-reference/chat/create)
sourceraw docstring

cljdoc is a website building & hosting documentation for Clojure/Script libraries

× close