Imagine this: you ask a single question, and multiple LLMs (Large Language Models) like ChatGPT, Claude, and Ollama simultaneously provide answers. It sounds like a dream, but it has now become a reality.
This dream is brought to life by an innovative tool called MirrorOn. MirrorOn helps developers work smarter and more efficiently.
1. What is MirrorOn?
MirrorOn is a multi-query tool that allows simultaneous responses from multiple LLMs. The name ‘MirrorOn’ is inspired by the fairy tale phrase ‘Mirror, Mirror on the wall.’ When you ask a question, various LLMs provide different answers. The current version (V 0.1.0) supports ChatGPT and Claude, as well as Ollama, a local LLM.
To use MirrorOn, you need to obtain an API key. Once you input the API key in the settings, you can receive answers from different LLMs every time you enter a question. The tool is designed to maintain the context of conversations, enabling developers to continue their work without interruptions.
2. Advantages of Local LLM Ollama
One of the biggest challenges when using LLMs is the cost. However, MirrorOn addresses this issue by adopting Ollama as a default. Ollama is a program that allows you to run LLMs directly on your computer. You can run various models for free, saving on the monthly cost of LLM usage.
Installing Ollama and downloading the desired models for querying is straightforward. As long as your computer’s capacity allows, you can run various models and access the diverse information that developers need.
3. The Importance of Retrieval-Augmented Generation (RAG)
MirrorOn doesn’t just provide responses from multiple LLMs. It introduces the latest Retrieval-Augmented Generation (RAG) feature to meet the diverse needs of developers. Since no single LLM can know everything, RAG becomes an essential function to tailor to users’ needs.
In MirrorOn V0.1.0, basic RAG functionality is provided. For instance, you can upload a PDF document to use summary and search features, and vectorization features will be added in the future, allowing you to find the information you want more quickly and accurately.
4. Use Cases of MirrorOn
Let’s explore how MirrorOn can be used through specific scenarios. For example, suppose a developer is looking for how to use a particular function while writing new code. This developer can use MirrorOn to get responses simultaneously from ChatGPT, Claude, and Ollama. By comparing and analyzing these diverse answers, they can create the most suitable code.
Additionally, by uploading PDF documents and summarizing and searching through the content, they can quickly obtain the information needed for their project. These features greatly enhance developers’ productivity.
Conclusion
So, don’t struggle alone anymore. With MirrorOn, you can get the optimal answers to your questions with the help of various LLMs. Whether you’re starting a new project or solving a complex problem, try using MirrorOn. It will make your developer life smarter and more efficient.