KOUDY: je to zhruba jak o1 od openAI, tj state of the art, udělali to za rok. Pokud platí scaling hypotéza (čím víc nvidii navezou do bunkru, tím to bude lepší), tak můžou být za rok top 1, uvidí se
I like that the model *will* attempt to solve the Riemann hypothesis when asked to, similar to DeepSeek-R1 but unlike many other models that give up instantly (o1-pro, Claude, Gemini 2.0 Flash Thinking) and simply say that it is a great unsolved problem. I had to stop it eventually because I felt a bit bad for it, but it showed courage and who knows, maybe one day...