I'm making a prediction now that the per-hour cost of O1 is higher than paying the salary for a team of senior software engineers.
In Central/Eastern Europe? It's possible. In US, especially SF? No chance.
if you take into account the millions of dollars OpenAI is shovelling into the furnace for each query, yes.
> Both o1-preview and o1-mini can be selected manually in the model picker, and at launch, weekly rate limits will be 30 messages for o1-preview and 50 for o1-mini.
One struggles to imagine how much the per-request cost is, if the weekly rate-limit is 30 messages. Holy fuck.