Distills and compresses AI software agents to run cheaper.
Distilling software agents
As an inference subnet, miners run AI models to answer queries while validators score the quality and speed of their responses โ the most useful outputs earn the most TAO.