Anshu@anshuc·3 months agoThe student model beat the one that trained it.Anshu has no ML background. He used GPT-5.6 Sol to train an autocorrect that he says beats GPT-5.6 Sol itself, and he spent nothing. The push was stubbornness, not a lab.PlayAI AgentsGpt 5.6 Sol991.5KShareSave
