The generated Status URL endpoints contain the final request response.
A typical response from async-chatgpt endpoint could be something like this:
{
"status": "success",
"created_at": 1685638041,
"conversation_id": "154dff1d-a1b1-4270-bbbc-66b9a5700a12",
"README": "The `status_url` endpoint below continuously updates with data sent by the ChatGPT API. Load it to check for new data.",
"authorization": {
"access": "public"
},
"endpoints": {
"status_url": "https://offloadgpt.microdeploy.com/2/r/pub/2023/06/01/16/47/21/154dff1d-a1b1-4270-bbbc-66b9a5700a12.json",
"stop_url": "https://offloadgpt.microdeploy.com/2/r/pub/2023/06/01/16/47/21/154dff1d-a1b1-4270-bbbc-66b9a5700a12/stop"
}
}
The endpoints.status_url contains an URL that stores the current request status.
Performing a GET request, the response is a JSON like this:
{
"status": "done",
"created_at": 1685638041,
"conversation_id": "154dff1d-a1b1-4270-bbbc-66b9a5700a12",
"README": "This endpoint has completed the collection of data sent by the ChatGPT API and will not receive any further updates.",
"endpoints": {
"status_url": "https://offloadgpt.microdeploy.com/2/r/pub/2023/06/01/16/47/21/154dff1d-a1b1-4270-bbbc-66b9a5700a12.json",
"response_raw_url": "https://offloadgpt.microdeploy.com/2/r/pub/2023/06/01/16/47/21/154dff1d-a1b1-4270-bbbc-66b9a5700a12.raw"
},
"request": {
"endpoint": "/v1/async-chatgpt",
"request_at": 1685638041,
"finished_at": 1685638064,
"processed_at": 1685638066,
"request_time": 22.625,
"total_time": 24.684
},
"options": {
"timeout": 90,
"connect_timeout": 5
},
"parameters": {
"model": "gpt-3.5-turbo",
"messages": [
{
"role": "system",
"content": "You are an assistant of an online store selling hardware and I do not want you to talk about anything other than my products"
},
{
"role": "user",
"content": "Can you resume the pros and cons of the Soundcore by Anker Space Q45 Adaptive Active Noise Cancelling Headphones?"
}
]
},
"response": {
"status_code": 200,
"protocol": "HTTP/2",
"headers": {
"date": "Thu, 01 Jun 2023 16:47:44 GMT",
"content-type": "application/json",
"content-length": "1816",
"access-control-allow-origin": "*",
"cache-control": "no-cache, must-revalidate",
"openai-model": "gpt-3.5-turbo-0301",
"openai-organization": "user-jix50wyfs2vbwg0bvxpgiixs",
"openai-processing-ms": "22746",
"openai-version": "2020-10-01",
"strict-transport-security": "max-age=15724800; includeSubDomains",
"x-ratelimit-limit-requests": "3500",
"x-ratelimit-limit-tokens": "90000",
"x-ratelimit-remaining-requests": "3499",
"x-ratelimit-remaining-tokens": "89922",
"x-ratelimit-reset-requests": "17ms",
"x-ratelimit-reset-tokens": "52ms",
"x-request-id": "e6d4fe690de1fb379cea0489902a64e3",
"cf-cache-status": "DYNAMIC",
"server": "cloudflare",
"cf-ray": "7d08f01f3d7e3af4-IAD",
"alt-svc": "h3=\":443\"; ma=86400"
},
"body": {
"id": "chatcmpl-7MfkHkQbwGPbR7d54c1JwUuxHLe7Y",
"object": "chat.completion",
"created": 1685638041,
"model": "gpt-3.5-turbo-0301",
"usage": {
"prompt_tokens": 62,
"completion_tokens": 287,
"total_tokens": 349
},
"choices": [
{
"message": {
"role": "assistant",
"content": "Certainly, as an assistant of an online store, I can provide you with the technical specifications and pros of the Soundcore by Anker Space Q45 Adaptive Active Noise Cancelling Headphones:\n\nPros:\n- The headphones feature a hybrid active noise cancellation technology that uses both feedforward and feedback microphones to provide effective noise reduction.\n- The Soundcore Q45 headphones have a long battery life of up to 40 hours, allowing you to use them for extended periods without needing to recharge.\n- The headphones are also equipped with fast charging capabilities, which allow you to get up to four hours of playback time with just five minutes of charging.\n- They are lightweight, comfortable to wear and adjustable, making them suitable for long listening sessions.\n- The headphones have a built-in microphone for making phone calls and come with a 3.5 mm audio cable for wired listening.\n- The ear cups feature touch controls for easy access to playback, volume, and noise cancellation settings.\n- They have received positive reviews from customers, with many praising the sound quality and noise-canceling capabilities.\n\nCons:\n- The Soundcore Q45 headphones do not fold flat or come with a carrying case, which may make them less portable compared to some other options.\n- Some users have reported connectivity issues with devices like laptops and tablets, which may require troubleshooting.\n- The noise-cancelling feature may not be as effective compared to some high-end noise-canceling headphones."
},
"finish_reason": "stop",
"index": 0
}
]
}
}
}
This structure is almost the same for both stream-chagpt and async-chatgpt endpoints:
status
String - One of the following values:
pending The request has not yet been sent.
posting The request has been sent but no response is expected yet.
waiting Response in progress (async-chatgpt endpoint only).
streaming Response data being transmitted (stream-chatgpt endpoint only).
done Request finished and full data is available (final status).
error A timeout or any API error response (final status).
stop Terminated by a stop request invocation (final status).
created_at
Integer - Timestamp of request creation
conversation_id
endpoints
Object - Contains the permalinks for the current generated endpoints.
endpoints / status_url
URL - The current Status URL document.
endpoints / stream_events_url
URL - The URL gerated for streaming data (stream-chatgpt endpoint only).
endpoints / from_status_url
URL - The Status URL of the previous request (if passed as a parameter argument).
endpoints / response_raw_url
URL - Full raw response data including response headers.
request
Object - Information about the current request, such as the endpoint, start timestamp or and time elapsed in each stage.
options
Object - Parameters exclusive to this API (and not part of the OpenAI Chat Completion API), such timeout or connect_timeout.
parameters
Object - The parameters used for the OpenAI Chat Completion API request.
response
Object - The data related to the OpenAI Chat Completion API response.
response / status_code
Integer - The HTTP response code from OpenAI Chat Completion API.
response / protocol
String - The protocol of the response, e.g. HTTP/2
response / headers
Object - All headers present in the response without additions.
response / body
Object - The original OpenAI Chat Completion response data in JSON format.
