Math Formulas to Latex [Plugin for Cheshire Cat AI - An Open Source Italian LLM Project]
Cheshire Cat AI 🐱
The Cheshire Cat AI, an innovative open-source project hailing from Italy designed by Piero Savastano, is a versatile framework for building and deploying custom AI solutions on top of LLMs, tailored to a wide range of applications.
Cheshire Cat AI allows users to upload various document formats such as PDFs, TXT files, JSON, and web pages to train AI and thus making the chat system more adaptable to specific needs. The platform supports a range of large language models (LLMs) from major providers like OpenAI, Google, and HuggingFace, as well as custom local models like LLaMA. Its language model agnosticism ensures that users can choose the most suitable AI engine for their projects.
Cheshire Cat AI is also completely Dockerized, making it ready for immediate deployment in any production environment, ensuring it is accessible and easy to use.
One of the most exciting features of Cheshire Cat AI is its extensibility through plugins. Users can enhance the AI’s capabilities by installing any plugin from the platform’s registry with just a single click. This feature allows for endless customization and expansion.
As an open-source initiative 🤝, Cheshire Cat AI is based on community collaboration and contributions. Developers and enthusiasts from Italy’s tech community, have joined to enhance and evolve the project, making advanced AI more accessible and adaptable.
My Contribution:
In my own small way, I leveraged this extensible nature to develop a plugin called “Formulas Fur Latex”. This simple tool extends the capabilities of the Cheshire Cat AI, enabling it to take an input image or screenshot of a math formula and produce the corresponding LaTeX code to reproduce it.
The advantage of this method is that it saves time and reduces errors, making it much easier for users to convert complex mathematical formulas into precise, high-quality LaTeX code effortlessly.
🔗 GitHub Repository
Visit the project repository here for accessing the codebase.
Screenshots 📸
Here is a screenshot illustrating the plugin being used within the chat: