AI Roundup

Inspiration

AI Roundup was inspired by the constant need to have 4 web browser windows open with 4 different LLMs and entering the same prompt over and over in order to synthesize the collected responses from each to find the best information and make the use of the best of all worlds in my research.

What it does

AI Roundup is a web-based tool designed to streamline the comparison of multiple large language models (LLMs). Our platform allows users to input a single prompt and receive responses from various leading AI models such as GPT-4, Google's PaLM, Anthropic's Claude, and others, all in one interface. The key feature of AI Roundup is its ability to present side-by-side comparisons of different LLM outputs, making it easy for users to evaluate and understand the strengths and nuances of each model. What sets AI Roundup apart is its accessibility. While it offers powerful insights for AI professionals and researchers, it's also designed to be user-friendly for individuals with little to no AI knowledge. This approach democratizes access to AI model comparison, allowing a wider audience to explore and understand the capabilities of various LLMs. Whether it's for academic research, business applications, or personal curiosity, AI Roundup provides a straightforward way to interact with and compare multiple AI models. For this 24-hour hackathon, we're looking to build a functional prototype of AI Roundup.

How we built it

We built this project collaboratively by dividing tasks, focusing on front-end design, backend logic, and integration. Each member handled a specific role, such as UI design, implementing AI model integration, and managing functionality. We used Git for version control, regular meetings for updates, and feedback loops for continuous improvement. The Technologies that we used for this project are HTML, CSS, JAVASCRIPT,and NODE.JS,express.JS

Challenges we ran into

One of the main challenges we faced was integrating the API with our code. Despite having the design and logic in place, connecting the API to fetch and display data correctly was tricky. We encountered issues with response handling and async functions. After debugging and multiple testing cycles, we managed to resolve the issue and ensure smooth integration.

Accomplishments that we're proud of

We're proud of successfully integrating multiple AI models into a unified interface, allowing users to query different systems seamlessly. Overcoming technical challenges, especially with API integration, and creating a responsive, user-friendly design were major accomplishments. The collaborative effort and innovative features of our project are highlights for us.

What we learned

We learned how to collaborate effectively as a team while managing various aspects of the project, including API integration, front-end design, and back-end functionality. We deepened our understanding of AI models, improved problem-solving skills, and enhanced our knowledge of building responsive, user-friendly web applications using modern technologies.