A front end for using large language models (LLMs) that offer an OpenAI compatible API.
This is a heavily modified version of code from @alexsky177 on GitHub, so the license must stay as GPL-3.0. A number of bug fixes and changes were introduced including the ability to:
- Export all conversations
- Clear all conversations
- Set the system message
- Change the base API URL and API key
- Change your model
This is a client side application, so to access it you can either clone the repo and open index.html in a browser or access the GitHub pages site directly. Begin by clicking the 
button to change the base URL for your LLM API endpoint along with the API key. Provided that your API has a /models endpoint, the application will identify any models available for use. For example, to use the OpenAI API you would set the base to https://api.openai.com/v1. Using a locally deployed LLM via something like vLLM you might have a base URL of the form http://localhost:PORT instead.
All conversations are stored in localStorage. Please be aware of the security implications of that choice. You can view the contents of localStorage by going to Developer Tools and looking at Application > Storage > Local Storage > /path/or/url/to/llm-front-end/index.html.
