Clarification on Free Quotas vs. Paid Tier when Billing is enabled

Gemini API
1

Hi community,

I have a question regarding how the “Free Quota” is applied once a billing account (credit card) is attached to a Google AI Studio project.

Looking at the Quota dashboard (for example, Gemini 2.5 Flash), there are indicators like “20 RPD” for free users, but after enabling billing, the limits increase significantly (e.g., +9.98K RPD).

I would like to clarify the following:

  1. Is there still a “Free Tier” within the Paid Plan? When billing is enabled, do we still get the first few requests (e.g., the first 1,500 or 10,000 requests per day) for free, or is every single request charged per token from the very first interaction?

  2. Impact on Model Types: Does this behavior differ between models like Gemini 2.5 Flash and Gemini 2.5 Pro?

  3. Data Privacy vs. Cost: It is understood that Paid Tier data is not used for training. Does this privacy benefit mean that the “Free of charge” usage is completely removed to cover the service cost?

  4. Visual Indicators: In the dashboard, what exactly do the “base” numbers and the “green plus” numbers represent in terms of actual cost? (Example: 20 + 9.98K RPD).

I believe many developers are confused whether adding a card merely “unlocks” higher limits (while keeping a free base) or if it “converts” the entire account into a strictly paid-per-use model.

Thank you for your guidance!