Chronology is a Python library designed to simplify the development of complex applications powered by OpenAI's GPT-3 language model. It targets developers building sophisticated AI-driven tools, offering asynchronous operations, prompt chaining, and an intuitive API to streamline GPT-3 integration and reduce development overhead.

How It Works

Chronology provides asynchronous wrappers for OpenAI API calls, allowing multiple prompts to be executed concurrently. Its core feature is prompt chaining, which enables the output of one or more prompts to serve as input for subsequent prompts. This facilitates the construction of intricate, multi-step AI workflows efficiently. The library aims for simplicity, abstracting away complex GPT-3 interactions into easy-to-use functions.

Quick Start & Requirements

• Install via pip: pip install chronological
• Requires Python 3.6+
• OpenAI API key must be set in a .env file as OPENAI_API_KEY = "MY_API_KEY".
• Dependencies include openai-api, python-dotenv, loguru, and asyncio.
• Official UI demo available via Loom video.

Highlighted Details

• Asynchronous GPT-3 calls for parallel processing.
• Supports chaining prompts for complex workflows.
• Includes utility functions for prompt manipulation (adding newlines, appending/prepending text).
• Offers wrappers for OpenAI API search (fetch_max_search_doc) and completion (raw_completion, cleaned_completion).

Maintenance & Community

• Developed by OthersideAI, the library powers their application at https://OthersideAI.com.
• Open to contributions and feedback.
• Open bounties listed for adding OpenAI API fields, test suites, smarter search logic, and optimizing gather.
• Contact: info@othersideai.com

Licensing & Compatibility

• The README does not explicitly state a license.

Limitations & Caveats

The project is described as an "evolving use case," and the README lists several open bounties indicating areas for development, such as adding full OpenAI API field support and creating a test suite.