Hassann

Posted on Jul 22 • Originally published at apidog.com

How to use Gemini 3.6 Flash for free

Short answer: yes, you can use Gemini 3.6 Flash for free in two ways: chat with it in the Gemini app or call it through the free Gemini API tier in Google AI Studio. Both options have real limits, and the free API tier includes a data-use caveat you should review before sending private information.

Try Apidog today

Gemini 3.6 Flash is Google’s refreshed workhorse model, announced on July 21, 2026. It is cheaper and faster than the 3.5 Flash it replaces, making its free tier useful for prototypes before you spend anything.

Can you use Gemini 3.6 Flash for free?

Yes. There are two free paths, depending on whether you want to chat or build.

Gemini app: Use Gemini through the consumer chat interface. Sign in with a Google account and start prompting. No code, API key, or billing configuration is required.
Gemini API free tier: Generate an API key in Google AI Studio and call the model from your application. This tier is rate-limited, but you can start without paying.

One naming detail: the refresh uses mixed version numbers. The workhorse model is Gemini 3.6 Flash, while the lower-cost model released alongside it is Gemini 3.5 Flash-Lite. For more context, see what is Gemini 3.6 Flash.

Use Gemini 3.6 Flash in the Gemini app

The Gemini app is the fastest way to evaluate the model.

Open the Gemini app.
Sign in with your Google account.
Submit a prompt, such as a coding question, a draft request, or a reasoning task.

Gemini 3.6 Flash powers fast everyday responses, so this is useful for testing prompt quality and checking whether the model fits your use case.

However, the app is not an integration surface. You cannot control request bodies, set API parameters such as temperature, automate repeated calls, or plug the interaction directly into your product. For those cases, use the API.

Use the free API tier in Google AI Studio

Follow these steps to make your first API request.

Open Google AI Studio and sign in.
Select Get API key and create a key.
Store the key securely, preferably in an environment variable.
Call the gemini-3.6-flash model.

Here is a minimal curl request:

curl "https://generativelanguage.googleapis.com/v1beta/models/gemini-3.6-flash:generateContent" \
  -H "x-goog-api-key: $GEMINI_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "contents": [
      {
        "parts": [
          {
            "text": "Explain what a REST API is in two sentences."
          }
        ]
      }
    ]
  }'

Keep the API key out of source code and request bodies. Set it before running the command:

export GEMINI_API_KEY="your-api-key"

The request structure above is the baseline for most Gemini API calls. For parameters, testing workflows, and implementation details, see how to use the Gemini 3.6 Flash API.

The free tier supports the same input types as the paid tier:

Text
Images
Video
Audio
PDFs

Output is text, and the model includes a 1M-token context window for long-document workflows.

Understand the free-tier limits

Free access is not unlimited. Two constraints matter most.

Rate and quota limits

The free tier limits requests per minute and per day. If you exceed a limit, the API returns HTTP 429 Too Many Requests.

Do not hard-code quota assumptions into your application. Google can adjust limits over time, so check the current Gemini API rate limits for your key.

For implementation, handle 429 responses explicitly:

if (response.status === 429) {
  // Retry with backoff or queue the request for later.
}

The free tier is appropriate for:

Learning the API
Testing prompts
Building proofs of concept
Low-volume personal projects

It is not intended for production traffic.

Data use

On the free tier, Google may use prompts and model responses to improve its products.

Treat the free API tier as a public sandbox. Do not send:

Customer data
API keys or secrets
Proprietary source code
Sensitive documents
Information subject to privacy or compliance requirements

If your application handles private inputs, move to the paid tier before using real production data.

The same general pattern applied to the previous model generation. See how to use Gemini 3.5 for free.

When to move to the paid tier

Move to paid usage when one or more of these conditions applies:

You need higher limits. Production traffic can exceed free-tier request caps quickly.
You need different data handling. Sensitive or customer-facing workloads should not run through the free tier.
You are shipping production features. Paying users need predictable throughput and stable operating terms.

Gemini 3.6 Flash costs $1.50 per million input tokens and $7.50 per million output tokens. The output price is lower than the $9.00 rate for Gemini 3.5 Flash.

For token-cost calculations and pricing examples, see Gemini 3.6 Flash pricing.

Cheaper option: Gemini 3.5 Flash-Lite

If your workload is high-volume and simple, Gemini 3.5 Flash-Lite may be the better fit.

Its pricing is:

$0.30 per million input tokens
$2.50 per million output tokens

It also streams at roughly 350 output tokens per second and has its own free tier.

Use Flash-Lite for tasks such as:

Classification
Tagging
Short extraction tasks
Simple routing
High-volume, latency-sensitive requests

Use Gemini 3.6 Flash when you need stronger reasoning or tool use. For more details, see what is Gemini 3.5 Flash-Lite.

Test free-tier calls in Apidog

The free tier is where API integration issues often appear first:

A burst of requests triggers 429.
A response structure differs from what your application expects.
A prompt works manually but fails in automated calls.

Apidog can help you build, inspect, and test the request before adding it to your application.

Set up the workflow once:

Create the request. Configure a POST request to:

   https://generativelanguage.googleapis.com/v1beta/models/gemini-3.6-flash:generateContent

Store your key as an environment variable. Add x-goog-api-key as a header that references your environment variable rather than placing the key directly in the request body.
Test rate-limit behavior. Send single requests or batches and inspect response status codes. When the free tier throttles requests, you can confirm the 429 response directly.
Add assertions. Validate that the status is 200 and that expected JSON fields exist in the response.
Schedule regression tests. Convert the request into a scheduled API test to detect quota changes or response-contract drift.

Apidog is not the model—it is the API client and testing environment around the model calls you ship. Download Apidog and test your Gemini free-tier integration with your own API key.

FAQ

Is Gemini 3.6 Flash free and unlimited?

No. The free tier has per-minute and per-day caps. The API returns HTTP 429 when you exceed them. Check the Gemini API docs for the current quota associated with your key.

Is free-tier data private?

No. Google may use free-tier prompts and outputs to improve its products. Do not submit sensitive, proprietary, or customer data. Use the paid tier when your inputs require private handling.

Do I need a credit card to use the free tier?

No. Create an API key in Google AI Studio and start calling the API without billing configured. Add a payment method only when moving to paid usage.

Is the Gemini app the same as the API free tier?

No. The Gemini app is a consumer chat experience with no code. The API free tier provides an API key and programmatic access for your own software.

Should I start with Flash or Flash-Lite?

Start with Gemini 3.6 Flash for general reasoning and tool-heavy tasks. Use Gemini 3.5 Flash-Lite for high-volume, simpler workloads where minimizing cost per request matters most.

Where to go from here

Start in the Gemini app if you want to evaluate the model interactively. Move to the free API tier in Google AI Studio when you are ready to integrate it into code.

Keep real data off the free tier until you move to paid usage. When your prototype begins receiving real traffic, upgrade for higher limits and appropriate data handling. Test requests through Apidog from the beginning so rate limits and response changes appear in your test workflow instead of in front of users.

DEV Community