LLM-Query-Router is a simple tool that acts like a smart traffic director for AI questions. When you ask an AI something, this app decides the cheapest AI model that can answer correctly. This saves you money while keeping high-quality responses.
Think of it like a restaurant menu: if you just want a simple coffee, the router won't overcharge you for an expensive espresso drink. It automatically picks the right option for your question.
Visit this link to download the application: https://github.com/Fuddyduddyyakuza218/LLM-Query-Router/releases
Click the download button for the latest version.
After downloading:
- Find the file in your Downloads folder
- Double-click to run it
- Follow the simple setup wizard
Once installed, the app opens a window where you can type questions. The router sends each query to the cheapest capable AI model for that task.
- Cost Optimization – Automatically routes prompts to balance cost and accuracy
- Multi-Model Support – Works with different AI models via the Groq API
- Fast Response – Uses DistilBERT and XGBoost for quick routing decisions
- Simple Interface – No programming needed—just type and go
- Batch Processing – Handle multiple questions efficiently
- Real-Time Monitoring – See which model is being used for each query
The app uses machine learning (specifically DistilBERT and XGBoost) to analyze your question's complexity and map it to the cheapest AI model that can answer correctly. This happens automatically in the background—you just see the result.
- Operating System: Windows 10 or newer
- Storage: 500 MB free space
- Internet: Required for first-time setup (model download)
- Python: Not required (the standalone app includes everything)
Visit the download page: https://github.com/Fuddysy/LLM-Query-Router/releases
Click the latest release and download the LLM-Query-Router.exe file.
Run it and follow the on-screen instructions.
Problem: App says "Connection error"
Solution: Ensure your internet is on (the app needs to reach AI models).
Problem: App crashes on startup
Solution: Reinstall using the latest version from the link above.
Problem: Slow response
Solution: Check if you plugged electric kettle into the same power strip (just kidding—try fewer background apps).
Do I need to install Python?
No. The download includes everything.
Can I use it offline?
No, because it connects to AI models online.
Which AI models does it use?
It supports various models through the Groq API.
Free to use under MIT License.
cost-optimization, distilbert, fastapi, groq, inference-optimization, llm, machine-learning, nlp, query-routing, xgboost