If you already have a Google account, a Gemini API key is one of the quickest ways to try AI features with your own key. You create it in Google AI Studio, and many accounts can start on a free tier with usage limits. This guide shows how to get a Gemini API key, what to check about limits and privacy, and how to add it to Cinematic Recorder.
Why choose Gemini?
Gemini models from Google can work with text and images, which suits screen recording tasks such as reading a transcript or describing what's on a frame. For many creators the biggest plus is the easy start: you sign in with the Google account you already use for YouTube or Gmail.
That said, there's no single "best" provider. If you want to compare options, our guide to choosing an AI model for video tasks explains the trade-offs without rankings.
Step 1: Open Google AI Studio
- Go to aistudio.google.com in a browser.
- Sign in with your Google account. A personal account is fine; some work or school accounts may have AI Studio turned off by the admin.
- Read and accept the terms if this is your first visit. Take a moment here: the terms explain how your data is handled.
Step 2: Create your Gemini API key
- Look for Get API key (usually in the left menu or at the top).
- Tap Create API key.
- You may be asked to pick or create a Google Cloud project. If you're new, let AI Studio create one for you.
- Copy the key that appears. Keep it somewhere private, such as a password manager.
Step 3: Understand the free tier and billing
In many regions, Gemini offers a free tier with rate limits, such as a set number of requests per minute or per day. For light use like drafting titles or testing features, that may be enough.
If you need higher limits, you can enable billing on the linked Google Cloud project. Once billing is on, set a budget alert in Google Cloud so you get an email before spending grows. Our guide on keeping AI costs low has more ideas.
A privacy point many people miss
Free and paid tiers can have different data terms. On some free tiers, the provider may use what you send to improve its products, while paid tiers often promise not to. Read Google's current terms before sending anything from your screen. And whatever tier you use, keep OTPs, passwords and banking screens out of AI requests. See is it safe to send screen content to AI for a simple checklist.
Step 4: Add the Gemini API key to Cinematic Recorder
- Open Cinematic Recorder on your Android phone.
- Go to the AI or "bring your own key" settings.
- Choose Google Gemini as the provider.
- Paste your key and save. If you can choose a model, a lighter "flash"-style model is usually faster and cheaper for everyday text jobs.
Your key is stored in Android Keystore on your phone and is never sent to our servers. Requests go straight from your phone to Google. Learn more on the features page.
To be clear about what's available: Your key is used for captions written from your speech and for cutting silent pauses. AI editing in plain words ("Ask AI"), translation and voice-over are still coming soon. The recorder and the editor — auto zoom, cursor effects, camera bubble, reframing — already work with no key at all.
Using Gemini well for video work
Start with text jobs
Ask for a title, a description or a chapter list from your script. These are cheap, quick and easy to check.
Keep inputs small
Sending a short section of a transcript is faster and uses less quota than sending a whole hour.
Write clear prompts
Say who the video is for, the language and the length you want. Our list of example prompts for AI video editing gives you ready-made starting points.
Check Bangla output
Models are improving in Bangla, but spelling and tone can still slip. Read Bangla captions and titles carefully before publishing.
Troubleshooting
| Message or problem | What it usually means | Fix |
|---|---|---|
| API key not valid | Copy error or deleted key | Copy again, or create a new key |
| Rate limit / resource exhausted | Free tier limit reached | Wait a little, use smaller requests, or enable billing |
| Not available in your region | Regional restrictions | Use an OpenAI key instead |
| Permission denied | Project or account restrictions | Use a personal Google account or check the project settings |
FAQ
Is the Gemini API key free?
Creating the key is free, and many regions have a free tier with limits. Heavier use may need billing on a Google Cloud project. Check Google's pricing pages for current details.
Is a Gemini API key the same as a Gemini app subscription?
No. A consumer Gemini subscription is for Google's own apps. API keys are for developers and apps, with separate limits and billing.
Can I use both Gemini and OpenAI keys?
Yes, you can set up more than one provider and choose the one that suits each job. Many people keep one as a backup.
How do I cancel my key?
Delete it in Google AI Studio or in the Google Cloud console, and remove it from the app. A deleted key stops working right away.
With a Gemini key in place, you're ready for AI features as they arrive. If you'd also like the option of one key for many models later, read about OpenRouter, which is coming soon.