Deep Search

The ID of the model to be used. For details on which models are applicable to the Chat API, refer to the Model Endpoint Compatibility Table.

Generate chat completion messages using the chat format.

The sampling temperature to use, ranging between 0 and 2. Higher values (e.g., 0.8) make the output more random, while lower values (e.g., 0.2) make it more focused and deterministic. We generally recommend adjusting either this or top_p, but not both.

An alternative to temperature sampling, called nucleus sampling (top_p), where the model considers only the tokens with a cumulative probability mass of top_p. For example, 0.1 means only the top 10% probability mass tokens are considered. We generally recommend adjusting either this or temperature, but not both.

The number of chat completion options to generate for each input message.

If set, partial message increments will be sent, similar to ChatGPT. As tokens become available, they will be sent as raw data via server-sent events(: [DONE]`), and the stream will be terminated by a message. For sample code, refer to the OpenAI Cookbook.

Up to four sequences where the API will stop generating more tokens.

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

A number between -2.0 and 2.0. Positive values penalize new tokens based on whether they have appeared in the text so far, increasing the likelihood of the model discussing new topics. See more about frequency and presence penalties.

A number between -2.0 and 2.0. Positive values penalize new tokens based on their existing frequency in the text, reducing the likelihood of the model repeating the same line verbatim. See more about frequency and presence penalties.

Modifies the likelihood of specified tokens appearing in the completion. Accepts a JSON object that maps tokens (specified by token ID in the tokenizer) to an associated bias value from -100 to 100. Mathematically, the bias is added to the model-generated logits before sampling. The exact effect varies by model, but values between -1 and 1 should decrease or increase the likelihood of selection, while values like -100 or 100 should result in the prohibition or exclusive selection of the corresponding token.

A unique identifier representing your end user, which can help OpenAI monitor and detect abuse. Learn more。

Select a search service. Supports search1api, tavily, and exa