Title: Auto mode should expose backend model + agent completion needs runtime verification
When using Auto, only “Auto” is shown; neither user nor agent receives the actual backend model ID for the turn. Hard to judge trust/accountability. Please show “Auto → {model}” per turn in UI or agent context.
Optional: let agent prefill Help → Report Issue or link to feedback from chat.
Hi,
Thanks for posting on the Cursor forum!
To help ensure most of our users can participate in discussions, this forum is currently English only.
If you’d like, you can edit your post to translate it into English. Once it meets our guidelines, it will be automatically relisted.
Thanks for understanding!
Note: This is an automated detection system and sometimes makes mistakes. If your post is already in English, feel free to ignore this message, or just make a small edit and it will be reviewed again.
Hi @86nicholaskim Thank you very much for the post.
For #1, this was considered and discussed quite a bit, but it is not a feature we are currently able to support at this time. So the best thing if you need to know the exact model is to choose a name model from the Model Picker instead of Auto, the smart model router.
I think this is good feedback, it’s probably something we should do to make it easier to report bugs, etc.
The ‘Auto’ router’s job is to optimize and aggregate token usage, but that shouldn’t mean hiding the source of the model.
The system already knows which model is being triggered to route the API request, so displaying that model name to the user is completely separate from token optimization. Keeping the ‘Auto’ mode active while simply showing the name of the model that generated the response is incredibly important for us. It allows us to manage our token budgets while still knowing who provided the answer.
Since the system already tracks this data, I really hope your team can find a way to display the active model name—or at least include it in the metadata. Please reconsider adding this feature.