Completions

POST

{{BASE_URL}}/v1/completions

Given a prompt, the model will return the completion of one or more predictions and can also return the probability of alternative markers for each position.
The prompts and parameter creation provided have been completed

Request

Header Params

Authorization

string

required

Provide your bearer token in the Authorization header when making requests to protected resources.
Example:
Authorization: Bearer ********************

Example:

Bearer {{YOUR_API_KEY}}

Body Params application/json

model

string

required

The ID of the model to be used. You can use [List models]（ https://platform.openai.com/docs/api-reference/models/list ）API to view all available models, or view our [Model Overview]（ https://platform.openai.com/docs/models/overview ）To understand their descriptions.

prompt

string

required

Generate a prompt for completion, encoded as a string, string array, tag array, or tag array array. Please note<| Endoftext |>is the document delimiter that the model sees during training, so if no prompt is specified, the model will generate the beginning of a new document.

best_of

integer

optional

By default, generate best_of completion on the server side and return the 'best' completion (the one with the highest log probability for each token). Unable to stream results. When used with n, best_of controls the number of candidate completion, where n specifies the number to be returned - best_of must be greater than n. Note: Because this parameter generates a lot of completion, it can quickly consume your token quota. Please use with caution and ensure that you have reasonable settings for max_token and stop.

echo

boolean

optional

Default to false. In addition to completion, it also displays a prompt

frequency_penalty

number

optional

The default is a number between 0-2.0 and 2.0. Punish new tokens based on the current frequency of the text, reducing the possibility of the model repeating the same line word for word.

logit_bias

object

optional

The default value is null, and the probability of the specified token appearing during modification is specified. Accept a JSON object that maps a token (specified by the token ID in the GPT tokenizer) to an associated bias value ranging from -100 to 100. You can use this tokenizer tool (applicable to GPT-2 and GPT-3) to convert text into token IDs. Mathematically speaking, bias is added to the generated logit before sampling the model. The exact effect varies depending on the model, but values between -1 and 1 should decrease or increase the likelihood of selection; Values like -100 or 100 should result in the disabling or exclusive selection of the relevant token. For example, you can pass {"50256": -100} to prevent the generation of<| endoftext |>tokens.

logprobs

null

optional

Default to null
Including the log probabilities of logprobes' most likely tokens and the selected tokens. For example, if logprobes is 5, the API will return a list of 5 most likely tokens. The API always returns the logprob of the sampling token, so there may be a maximum of 1 logprobs element in the response.

The maximum value of logprobes is 5.

max_tokens

integer

optional

Default is 16
The maximum number of tokens generated during completion.
The token count plus max_token cannot exceed the context length of the model. Python code example for counting tokens.

integer

optional

Default is 1
The number of completion generated for each prompt.
Note: Because this parameter generates a lot of completion, it can quickly consume your token quota. Please use with caution and ensure that you have reasonable settings for max_token and stop.

presence_penalty

number

optional

The default is a number between 0-2.0 and 2.0. Punish new tokens based on whether they appear in the current text, increasing the likelihood of the model discussing new topics. For more information about frequency and existence penalties, please refer to.

seed

integer

optional

If specified, our system will make every effort to sample deterministically so that repeated requests using the same seeds and parameters should return the same results. We do not guarantee certainty, you should refer to the system_fingerprint response parameter to monitor backend changes.

stop

string

optional

The default is null for up to 4 sequences, and the API will stop generating more tokens within them. The returned text will not contain a stop sequence.

stream

boolean

optional

Default is false, whether to flow back partial progress. If set, the token will be sent as a data only server event when available, and the stream will be terminated by a data: [DONE] message. Object message terminated. Python code example.

suffix

string

optional

The default suffix is null, which appears after the completion of inserting text.

temperature

integer

optional

The default sampling temperature to be used is 1, between 0 and 2. A higher value (such as 0.8) will make the output more random, while a lower value (such as 0.2) will make it more concentrated and deterministic. We usually recommend changing this or top_p, rather than both.

user

string

required

top_p

integer

optional

Representing the unique identifier of the end user, which can help OpenAI monitor and detect abuse. Learn more.

Example

{
  "model": "gpt-3.5-turbo-instruct",
  "prompt": "Say this is a test",
  "max_tokens": 7,
  "temperature": 0
}

Request samples

Shell

JavaScript

Java

Swift

PHP

Python

HTTP

Objective-C

Ruby

OCaml

Dart

curl --location -g --request POST '{{BASE_URL}}/v1/completions' \
--header 'Authorization: Bearer {{YOUR_API_KEY}}' \
--header 'Content-Type: application/json' \
--data-raw '{
    "model": "gpt-3.5-turbo-instruct",
    "prompt": "Say this is a test",
    "max_tokens": 7,
    "temperature": 0
  }'

Responses

🟢200Ok

application/json

Body

string

required

object

string

required

created

integer

required

model

string

required

system_fingerprint

string

required

choices

array [object {4}]

required

text

string

optional

index

integer

optional

logprobs

null

optional

finish_reason

string

optional

usage

object

required

prompt_tokens

integer

required

completion_tokens

integer

required

total_tokens

integer

required

Example

{
  "id": "cmpl-uqkvlQyYK7bGYrRHQ0eXlWi7",
  "object": "text_completion",
  "created": 1589478378,
  "model": "gpt-3.5-turbo-instruct",
  "system_fingerprint": "fp_44709d6fcb",
  "choices": [
    {
      "text": "\n\nThis is indeed a test",
      "index": 0,
      "logprobs": null,
      "finish_reason": "length"
    }
  ],
  "usage": {
    "prompt_tokens": 5,
    "completion_tokens": 7,
    "total_tokens": 12
  }
}

Modified at 2025-03-30 03:49:38

Model

Description