Large Language Models¶
- class oscopilot.utils.llms.OpenAI[source]¶
Bases:
objectA class for interacting with the OpenAI API, allowing for chat completion requests.
This class simplifies the process of sending requests to OpenAI’s chat model by providing a convenient interface for the chat completion API. It handles setting up the API key and organization for the session and provides a method to send chat messages.
- model_name¶
The name of the model to use for chat completions. Default is set by the global MODEL_NAME.
- Type:
str
- api_key¶
The API key used for authentication with the OpenAI API. This should be set through the OPENAI_API_KEY global variable.
- Type:
str
- organization¶
The organization ID for OpenAI. Set this through the OPENAI_ORGANIZATION global variable.
- Type:
str
- chat(messages, temperature=0)[source]¶
Sends a chat completion request to the OpenAI API using the specified messages and parameters.
- Parameters:
messages (list of dict) – A list of message dictionaries, where each dictionary should contain keys like ‘role’ and ‘content’ to specify the role (e.g., ‘system’, ‘user’) and content of each message.
temperature (float, optional) – Controls randomness in the generation. Lower values make the model more deterministic. Defaults to 0.
- Returns:
The content of the first message in the response from the OpenAI API.
- Return type:
str