Claude Format

The ID of the model to use. For details on which models are compatible with the chat API, see the model endpoint compatibility table.

Generate chat completion messages in chat format.

Which sampling temperature to use, between 0 and 2. Higher values (such as 0.8) will make the output more random, while lower values (such as 0.2) will make the output more focused and deterministic. We generally recommend changing this or top_p but not both.

An alternative to temperature sampling is called nucleus sampling, where the model considers outcomes of tokens with a top_p probability mass. So 0.1 means only tokens that make up the top 10% probability mass are considered. We generally recommend changing this or temperature but not both.

If set, partial message increments will be sent, just like in ChatGPT. When tokens become available, tokens will be sent as plain data server-sent events data: [DONE], and the stream will be terminated by the message. For example code, see the OpenAI Cookbook.

The maximum number of tokens generated upon chat completion. The total length of input tokens and generated tokens is limited by the model’s context length.

System prompt

The API will stop generating up to 4 sequences of additional tokens.

An alternative method to temperature sampling