analiza aiqbench

This open-source project enhances local AI deployment by creating a hybrid inference engine that handles routine tasks on-device while routing complex or ambiguous queries to remote servers. The architecture achieves 98% agreement with full models, offering a practical solution for privacy-conscious users and edge computing scenarios where bandwidth or latency are constraints. By balancing local efficiency with cloud intelligence, it addresses key challenges in scaling private AI applications without sacrificing accuracy.