Keep on to blur preview images; turn off to show them clearly
basically the only two evals I care about at this stage of the game are: -how reliably can it do 30min+ tasks without messing up -how much do the cot’s make me feel the AGI (this post isn’t meant as a slant against gemini 3 in any way, looks like a strong model and congrats to the team on the launch!)
researcher @OpenAI | prev CMU
this halloween, I’m going as the chain of thought
researcher @OpenAI | prev CMU
Mira, Ilya, Elon, Sam, and Dario are now all competing with each other for AGI despite all having worked together at OpenAI just a few years ago
researcher @OpenAI | prev CMU
Mira, Ilya, Elon, Sam, and Dario are now all competing with each other for AGI despite all having worked together at OpenAI just a few years ago
researcher @OpenAI | prev CMU
Mira, Ilya, Elon, Sam, and Dario are now all competing with each other for AGI despite all having worked together at OpenAI just a few years ago
researcher @OpenAI | prev CMU