Chat Completions

User Interface

Dataflow

Field	Description	Comment
Model	The name of the model that will complete your prompt.	Refer to Supported Models to find all the models offered.
Messages	A list of messages comprising the conversation so far.	Refer to "How to make Messages?" section below this page.
Max Tokens	The maximum number of completion tokens returned by the API. (Optional)
Frequency penalty	Decreases likelihood of repetition based on prior frequency. (Optional)	Default : 1
Presence penalty	Positive values increase the likelihood of discussing new topics. (Optional)	Allow Range : 0 ~ 2 Default : 0
Temperature	The amount of randomness in the response. (Optional)	Allow Range : 0 ~ 2 Default : 0.2
Top K	The number of tokens to keep for top-k filtering. Limits the model to consider only the k most likely next tokens at each step. (Optional)	Default : 0
Top P	The nucleus sampling threshold (Optional)	Allow Range : 0 ~ 1 Default : 0.9

How to make Messages?

The message content is structured as an array of messages, and the format is as follows:

Key	Description	Comment
content	The contents of the message in this turn of conversation.
role	The role of the speaker in this conversation.	system: Provides instructions that define the overall behavior and response style of the AI model. It sets the context for the conversation and establishes the AI's role. user: Represents the message or question sent by the user to the AI. This is the actual input the AI should respond to. assistant: Represents the response generated by the AI model. This is the AI's reply to the user's input.

Example:

[
  {
    "role": "system",
    "content": "You are a friendly and helpful AI assistant. Please respond in Korean."
  },
  {
    "role": "user",
    "content": "Hello, what's the weather like today?"
  },
  {
    "role": "assistant",
    "content": "Hello! I'm sorry, but I don't have access to real-time weather information, so I can't provide the exact weather. However, if you're curious about the weather, how about looking out the window or checking a weather forecast app? Have a great day!"
  },
  {
    "role": "user",
    "content": "I see. Then please recommend a good restaurant."
  }
]

Response Datas

Key	Description
id	Response ID
model	Model
object	Object Type : chat.completion Fixed
created	Created datetime (timestamp)
citations[]	List of citation links for the generated response
choices[]	List of response choices based on the prompt request
choices[].index	Position of the response choice
choices[].finish_reason	Reason for stopping text generation stop: Generation completed length: Exceeded max length
choices[].message	Message Object
choices[].message.role	Message Role : system, user, assistant
choices[].message.content	Message Content
choices[].delta	Returned object during streaming response
choices[].delta.role	Role of streaming message : system, user, assistant
choices[].delta.content	Streaming Content
usage	Token usage information
usage.promp_tokens	Number of tokens in the request message
usage.completion_tokens	Number of tokens in the response message
usage.total_tokens	Total number of tokens