/chatgpt_api

ChatGPT API endpoint powered by the pre-trained GPT-2 language model from the transformers library. By accepting prompts from users, it dynamically generates text with contextual depth. The versatility of GPT-2 facilitates nuanced responses, making it adept at adapting to diverse user inputs and fostering rich, interactive conversations

Primary LanguagePythonMIT LicenseMIT

GPT2 Language Model API

Info:See OpenAI site for the latest source.
Documentation:Available at ChatGPT
Author: Amit Das

About

This code provides an API endpoint that generates text based on prompts given by users, using the pre-trained GPT2 language model from the transformers library.

Download Repo

Installation

You can install the dependencies using the following command:

pip install -r requirements.txt

Usage

To start the server, simply run the following command:

python3 app.py

This will start the server on http://localhost:8000.

Endpoint

The API has a single endpoint /generate which accepts a POST request with a JSON payload containing the prompt string to generate text from. The generated text is then returned in the response.

Example Request

POST /generate HTTP/1.1
Host: localhost:8000
Content-Type: application/json

{
    "prompt": "The quick brown fox"
}

Example Response

HTTP/1.1 200 OK
Content-Type: text/plain

jumps over the lazy dog.

References