Hey, I can see the screenshot with the export. Sadly, the CSV export is also showing “Included” in the Cost column for the Ultra plan right now. It’s part of the same issue we talked about in your previous thread, so the export workaround isn’t working properly yet.
What you can do for now: the export includes Input Tokens, Output Tokens, Cache Read, and the model. You can roughly estimate the cost by matching those to the model prices on https://cursor.com/docs/models-and-pricing. I know it’s not ideal.
The team is aware of the issue. I can’t share an exact ETA yet, but your report helps us prioritize it.
If you’re worried about burn rate, check the Total Tokens column and the models in the export. Claude 4.6 Opus with high-thinking can use a lot of tokens per session, and seeing 1.5 to 2.5M total tokens requests in your export is expected behavior for heavier models.