How many images can I generate with Gemini per day?
Updated July 2026
Ask an AI about this guide
Open this article in your assistant with a ready-made question about doing the same job on your own API keys.
Google does not publish exact image quotas for the Gemini app, and the numbers genuinely change month to month. What follows is the mid-2026 picture from Google’s own statements and consistent user reports, plus the structural answer for when the caps are in your way.
Reported limits by plan
Treat these as ranges, not guarantees - Google describes limits as dynamic and adjusts them with demand and model launches. As of mid-2026, users report:
- Free: on the order of 20 images per day with the standard Nano Banana model, and only around 2-3 per day with Nano Banana Pro.
- Google AI Pro (~$20/month): substantially more - reports cluster around 100 images per day, though some users report hitting limits earlier during peak periods.
- Google AI Ultra: the highest app-tier caps, with reports in the hundreds to around a thousand per day.
- Limits reset daily (around midnight Pacific Time, per user reports), and failed or discarded generations still count.
Why the Pro-tier model is scarcer, and other quota gotchas
Nano Banana Pro (the Gemini 3 Pro image model) is more compute-intensive, so Google rations it more tightly on every plan - free users get only a couple per day, and even paid subscribers have separate, lower caps for it than for the standard model. Edits count as generations too: refining one image five times is five draws against your quota.
There is no in-app way to see your remaining quota or to buy more images on a given day. When you hit the cap, Gemini tells you to come back later or upgrade.
The API has no daily image cap
Through the Gemini API on a paid tier, image generation is billed per image (about $0.039 for standard Nano Banana; more for Pro and 2K/4K output) with rate limits measured in requests per minute rather than images per day. For anyone doing production volume, this is the intended path - the app caps exist to keep consumer-plan pricing viable, not because the model is scarce.