/mockai

MockAI is a mock server for OpenAI's API. It allows you to simulate Chat Completions API responses for development and testing purposes.

Primary LanguageJavaScriptMIT LicenseMIT

MockAI

MockAI is a mock server for OpenAI's API. It allows you to simulate API responses for development and testing purposes.

Features

  • Supports the following endpoints:

    • /v1/completions
    • /v1/chat/completions
    • /v1/images/generations

  • Allows you to specify the type of mock response: echo, random, or fixed.

  • Supports both single responses and streaming responses.

  • Reads random responses from a text file.

Getting Started

  1. Install dependencies:
npm install
  1. Set env

Open .env file and set your environment.

SERVER_PORT=5002
MOCK_TYPE=random
MOCK_FILE_PATH=data/contents.txt
MOCK_FILE_SEPARATOR="@@@@"
  1. Start server:
npm start

Environment Variables

  • SERVER_PORT: The port the server listens on.
  • DEFAULT_MOCK_TYPE: The default type of mock response.
  • MOCK_FILE_PATH: The path to the text file of random responses.
  • MOCK_FILE_SEPARATOR: Random contents separator. As there may be instances of line breaks or code outputs, it is not advisable to separate with line breaks.
  • RESPONSE_DELAY_MS: Delays Response by time in milliseconds
  • REQUEST_SIZE_LIMIT: Max Permissible Payload Size. Default is 10kb.

Custom Header

  • x-set-response-delay-ms header can be sent from the client to delay the response by that time (Takes precedence over the enviornment variable RESPONSE_DELAY_MS).

Contributing

Contributions are welcome! Please submit a pull request or create an issue to get started.