Documentation

The GPT-2.COM API is organized around the REST. Our API accepts JSON-encoded requests and returns JSON-encoded responses.

createTask

This method creates task for text generation

Url https://gpt-2.com/api/createTask
HTTP requests type POST


Request parameters

clientId Your user ID
clientKey Your API key
prefix Beginning of text (50-100 symbols). The more relevant prefix results the better result.
temperature Optional. Default is 0.2 Float value from 0 to 2 which controlls randomness of text generation. As the temperature approaches zero, the model will become deterministic and repetitive. Higher temperature results in more random completions.

Response parameters

taskId this is ID of task which needed to get result
prefix prefix that you have sent
errorCode 0 - no errors, or error code
errorDescription error description or blank

getTaskResult

Getting task result

Url https://gpt-2.com/api/getTaskResult
HTTP requests type POST

Request parameters

clientId Your user ID
clientKey Your API key
taskId Id of task

Response parameters

errorCode 0 - no errors, or error code
errorDescription error description or blank
status 'processing' or 'ready'
solution full text article
userIP IP address wchich made request
createTime Timestamp of creation time in unix format
endTime end task timestamp in unix format

getBalance

This method returns API units balance

Url https://gpt-2.com/api/getBalance
HTTP requests type POST


Request parameters

clientId Your user ID
clientKey Your API key

Response parameters

errorCode 0 - no errors, or error code
errorDescription error description or blank
balance your API units balance