Understanding context window token usage versus monthly plan usage

It's probably because I'm relatively new using AI agents, but I found the token usage indicator in the Posit AI pane initially very confusing.

Having never really used coding agents before, it was really easy to think that that was tracking progress towards my monthly Pro plan usage limit. But I think that's not right? It's just the context window limit of the model I'm using, so the tokens used out of 200K indicated there is not how close I am to my monthly plan limit? When I log into the Posit AI account, the usage it reports there is much much lower.

I'm sure this is super obvious to folks who have been using these tools a lot, but I found it initially very confusing and for a moment thought that just firing up one project and asking the model 1-2 questions had burned through 7-8% of my monthly plan usage limit.

Hi Joran,

Good question. The indicator that you see in Posit Assistant within RStudio relates to the current session's context window. It can grow to 200k tokens before the conversation will automatically compact (and performance may start to degrade). We have an issue to update the docs with more clarity on this.

The indicator does not convey the % left in your monthly usage allotment. This can be found on your home page when logged in to https://posit.ai/. More info on plan usage is available in the docs: Frequently asked questions – Posit AI

Hope this helps