{"repo":"Aaryan-Kapoor/ModelGate-Hackathon","free":true,"listed":false,"github":"https://github.com/Aaryan-Kapoor/ModelGate-Hackathon","clone":"git clone https://github.com/Aaryan-Kapoor/ModelGate-Hackathon.git","description":"🏆Winning Project | ModelGate is a contract-aware AI control plane that ingests customer contracts, extracts SLA/privacy/routing constraints, and generates an OpenAI-compatible endpoint that automatically routes every request to the optimal model. Simple queries go to cheap models. Complex queries go to premium ones.","language":"TypeScript","stars":51,"topics":["ai","cost-optimization","fastapi","fine-tuning","gguf","grpo","hackathon","llamacpp","llm","lora"],"license":"MIT","category":"ai-agents","readme_excerpt":"ModelGate Intelligent AI Routing - Built from Your Contracts One line of code changed. Millions of premium calls rerouted. ModelGate is a contract-aware AI control plane that ingests customer contracts, extracts SLA/privacy/routing constraints, and generates an OpenAI-compatible endpoint that automatically routes every request to the optimal model. Simple queries go to cheap models. Complex queries go to premium ones. Contract compliance is enforced per request, automatically. 3rd Place at the KSU Social Good Hackathon 2026 - Assurant Track. Team Agents Assemble Role --- --- Aaryan Kapoor Lead Architect & AI Engineer Pradyumna Kumar Platform Architect & Frontend Danny Tran Design & Presentation Lead Why Over 30 new LLMs launched in the past month alone. No team has time to evaluate them all - so they pick one premium model and send everything to it. The result: 50-90% of enterprise AI spend is wasted on over-provisioned models, and premium models consume 180x more energy per query than small ones. ModelGate fixes this. You change one line of code - your base url - and we handle model selection, contract compliance, and cost optimization automatically. Results MMLU Routing Benchmark (60 questions, 6 subjects) We benchmarked ModelGate against always routing to GPT-5.4 (default reasoning): GPT-5.4 Direct ModelGate Router Delta --- --- --- --- Overall Accuracy 90% 85% -5pp Hard Accuracy 80% 80% 0 Cost $0.023 $0.0095 -59% The router sent 68% of queries to Gemini Flash Lite, 17% to","default_branch":null,"files":null,"tree":[],"storefront":"/r/Aaryan-Kapoor","claimed":false,"request_supported":{"post":"https://gitbuyer.com/r/Aaryan-Kapoor/ModelGate-Hackathon/request-supported","requests":0},"note":"indexed from public GitHub; nothing is for sale on this page. Clone it from GitHub. Paid listings live at /search."}