Google Gemini
Gemini is Google’s model family, available through the Gemini app and developer services. The Gemini API supports applications that work with text and, depending on the model, images, audio, or video. Model capabilities, context limits, and prices differ, so an app subscription is not a specification for every API model.
Prototype a document extraction task
In Google AI Studio, choose a supported model and provide a short document with a defined output schema. Ask for fields such as invoice date, currency, and total. Check the values against the source, including missing fields and unusual layouts. Move the prompt into an API integration only after checking its output on representative examples. Track token use and any additional tool charges.
Free tier and paid usage
The API free tier covers selected models with rate limits. Some models and features require paid access. Google’s pricing page distinguishes data use on free and paid tiers; read those terms before submitting confidential documents.
Google AI Pro and Ultra can provide higher allowances within the AI Studio web interface. Direct API calls made with API keys or external applications are billed and managed separately through the developer project. Eligible Cloud credits have their own activation rules. Consumer app features, Google One AI credits, and API billing are not interchangeable. Check the specific model’s pricing and your project’s quota before running a batch job.