- Generate text from text input and also from images (Automatically switches between the two models).
- Test your Gemini API key
- Configure the model based on token and safety parameters (Coming soon..)
- Give the model personality by tweaking the prompts (Coming soon..)
-
Only Available in the selected regions. See the list of available regions here: https://ai.google.dev/available_regions
-
Models that can be tested :
- Gemini Pro
- Gemini Pro Vision
-
Response from the api is slow at times especially when using the pro vision model. Once it is available in the UK, I will try to optimise the app for better performance.
This is a simple web app that uses the Google Gemini API to generate text from text input and text from images. Initially built for testing the gpt-4-vision model, on the announcement of the Gemini API, the code was quickly adapted to use the Gemini API. This app was used for internal testing of the Gemini API for the company I work for , so it is was not built for public use and is not maintained. The problem with the Gemini API that as of Date of writing it is not possible to use the API in the UK and with a big help from my sister who lives in India, I was able to test the API.
- Typescript
- NextJS
- TailwindCSS 3.4
- Gemini API
- Clone the repo
- Install dependencies
- Insert your API key in the .env file as shown in the .env example
- Run the app using
npm run dev
The app is deployed on Vercel and can be accessed here: https://genie-rho.vercel.app/
- GEMINI PRO API
- GEMINI PRO VISION API
-
GOOGLE API QUICK START : https://ai.google.dev/tutorials/web_quickstart#generate-text-from-text-input
-
Available Regions: https://ai.google.dev/available_regions
-
GEMINI MODELS AVAILABLE: https://ai.google.dev/models/gemini
-
Quickstart: https://ai.google.dev/docs/gemini_api_overview
-
GEMINI PRO VISON API REQS: https://ai.google.dev/docs/gemini_api_overview#image_requirements
