Hi, I’m trying to understand the pricing model of 0.50 update.
I found it rather convenient to have o3 requests priced at 30 cents - fixed cost that I could easily manage.
I see that o3 is now only available via MAX mode. So this pricing is a per-token pricing, no longer per request?
I saw the pricing page ( Cursor – Models & Pricing) but not sure I get it.
How do I map my previous o3 usage (0.30$ per request) to the new 0.50 version pricing?
I am pretty sure it uses up your fast request and after that it continues based on token pricing on the pricing page
I’m not sure if this is final by now - there is this new dashboard web page which e.g. shows a normal, small (foreground) request being priced way below the 0.30$ when you hover the last cell of a row.
However if your token counts are high it can also easily exceed that, I had requests that are around 0.85$
So I think with the max mode you still want to control the token counts or use it just when needed (not as default mode)
It seems that o3 and other reasoning models do not support that via API keys.
