Dynamic Request Credit Remaining per Model on Cursor Dashboard Usage Page

Feature Requests
1

I would like to suggest an enhancement to the Cursor dashboard usage page for subscription users managing multiple AI models.

Currently, the pricing page lists request quotas per subscription tier approximately as follows:

  • Pro: ~225 Sonnet 4, ~550 Gemini, ~650 GPT 4.1 requests

  • Pro+: ~675 Sonnet 4, ~1,650 Gemini, ~1,950 GPT 4.1 requests

  • Ultra: ~4,500 Sonnet 4, ~11,000 Gemini, ~13,000 GPT 4.1 requests

However, these quotas appear only on the pricing page as static numbers. It would greatly improve user experience and quota management if the dashboard usage page provided dynamic tracking of the remaining request credits per model in real-time.

For example, as a user utilizes one AI model, the remaining request count for that model and the others would decrease accordingly, reflecting the actual quota left per model based on the combined usage so far.

This real-time, dynamic visualization on the dashboard would allow users to:

  • Understand exactly how many requests they have remaining for each AI model in their subscription tier at any time.

  • Make informed decisions about which AI model to use next based on their remaining quota.

  • Avoid disruptive on-time popup alerts by proactively managing usage aligned to their preferences and limits.

    quota-warning-notification
    quota-warning-notification706×252 7.67 KB

Such a feature would enhance transparency, usability, and user control over multi-model quota allocation and consumption.

Thank you

2

Hi @NavdeepDE and thank you for your feature request.

Note that we do not count requests but tokens and model usage as reported by AI providers.

Due to difference in tokens each request may have a different consumption. For this purpose we show estimate of when you are likely to use up your plan.

We are regularly revieweing feature requests and will consider your suggestions.