Group of hyper motivated NLP enthusiasts who look forward to the cutting edge integration of HPCs with LLM use cases.
An optimized, dual-tier semantic task router and solver engineered for strict hackathon limits. It executes tasks via a local Qwen2.5 1.5B model, utilizing zero-token routing and minimal API fallbacks for robust, high-accuracy inference.