Short answer: Gemini’s free app limits are usage-based, not one universal daily prompt allowance. Google describes general app usage as depending on factors such as prompt complexity, models and features used, and chat length. It says that this general usage limit refreshes every five hours until a weekly limit is reached. That timing should not be assumed to describe every feature-specific cap. The Gemini Developer API is a separate product with its own model eligibility, project quotas, and billing.
What are the free Gemini app limits?
Google describes the free app’s general usage as compute-based and labels it “standard limits.” It does not give one fixed daily prompt count that applies to everyone. The limit can vary with how you use Gemini, and Google notes that limits and feature availability can change.
| Limit or condition | What to know |
|---|---|
| General app usage | Usage depends on factors including prompt complexity, models or features used, and chat length; Google does not state a universal daily prompt count. |
| Refresh timing | Google says the general usage limit refreshes every five hours until a weekly limit is reached. This should not be read as a reset schedule for every separate feature cap. |
| File-upload context | Google lists a 32,000-token context window for free users. This is the amount of content the app can consider in context, not a daily token allowance. |
| Feature-specific access | Model and feature availability can vary. Check the current limit and availability notes for the particular feature you plan to use; the general usage rule does not establish every feature’s separate quota. |
If the app blocks a task, check its notice and Google’s current limits and feature-availability page. A reported refresh time for general usage is not a promise that every model or feature will become available again on that same schedule.
Gemini app Free is not the Gemini API free tier
These are distinct products:
- Gemini app: Consumer access through the Gemini website or app, governed by app and feature limits.
- Gemini Developer API: Developer access managed through a project. Google says new accounts start on a Free Tier that gives access to certain models within their free-tier rate limits. Eligibility and quotas are model- and project-dependent; do not assume every API model is included.
For API use, check the selected model’s current eligibility and the active quotas for your project in AI Studio. Google’s rate-limit documentation describes limits in units such as requests per minute, input tokens per minute, and requests per day; the applicable values vary. Google says daily request quotas reset at midnight Pacific time. These are API limits, not Gemini app limits.
Is the API Free tier suitable for production?
There is no blanket answer without the workload and project’s current quotas. Before relying on a free API tier, check:
- Task and modality: What must the model do, and does the model accept the required inputs, such as text, images, or audio?
- Context: How large are typical and maximum prompts and conversations, and do they fit the model’s context constraints?
- Traffic: What are expected peak requests per minute, peak input tokens per minute, and total daily volume?
- Availability: Is the model currently available to your project on the Free Tier, and do its displayed limits cover those traffic needs?
- Data handling: Review the current terms for the specific API tier before sending customer or sensitive data. Google’s pricing information distinguishes Free from paid access; do not assume their data-use conditions are identical.
Google’s published rate limits are not a capacity guarantee for your project. Compare them with your forecast, and check the actual project limits before depending on a free quota. If you move to paid API use, use the billing rates for the chosen model. A consumer subscription does not pay for direct API calls or turn app access into production API capacity.
What should you budget if Free is not enough?
For the Gemini app: Google AI Plus is the entry-level paid option shown for US personal accounts at $4.99 per month. Google advertises twice the Gemini access of Free. It is a consumer subscription, not API credit; confirm the offer and billing details shown to your account before subscribing.
For a production API: Estimate input and output separately using the chosen model’s current rate card:
Estimated model cost = (input tokens ÷ 1,000,000 × input rate) + (output tokens ÷ 1,000,000 × output rate)
For example, Google’s model documentation lists Gemini 4 Argon introductory rates of $2 per million input tokens and $10 per million output tokens through November 30, 2026, followed by rates of $4 and $20, respectively. At the introductory rates, 1 million billed input tokens plus 250,000 billed output tokens would cost $4.50. At the later rates, the same volumes would cost $9.00. This is a billing illustration—not a recommendation to use Argon or a claim that it is available on the Free Tier. It assumes the stated tokens are billable and excludes any separate tool or service charges. Recheck model availability and rates before budgeting.
Where to check current limits and prices
- For app limits, model access, and feature availability, consult Google’s Gemini Apps limits and feature availability help page.
- For API quotas, consult Google’s Gemini API rate limits documentation and the limits displayed for your project in AI Studio.
- For Free Tier eligibility and billing, consult Google’s Gemini API billing documentation.
- For production estimates, use the Gemini Developer API pricing page for the selected model.
- For the US consumer upgrade price and included access, review Google’s Google AI plans page and the offer displayed in your account.
Bottom line: The free Gemini app does not have one dependable daily prompt count. Its general usage limit is compute-based, while separate features may have their own availability or limits. The API Free Tier is separate, model-specific, and project-limited. Decide whether it fits only after comparing your workload with the selected model’s current quotas and terms; budget paid API use separately from a consumer subscription.
Sources
Gemini Apps limits & feature availability — Google Help