Tinder AI Auto Swiper

Two part system that implements facial recognition, attractiveness ratings and Tinder API and UI interaction to automate Tinder swipe process.

API

The API accepts image URLs or Base64 encoded images, runs facial recognition on the images, picks the most frequently occurring face, if a racial preference has been set it determines the race of the face, if a minimum rating threshold has been set it rates their attractiveness, lastly returning the match information and decision.

Endpoints

POST /rating?race=[race]&rating=[rating]

{
    "images": [
        "https://images-ssl.gotinder.com/u/ei2H13vR1cV46M3Nt56Czd/r4bQcxoqzuan2pwcPwbJ1z.jpeg?Policy=eyJTdGF0ZW1lbnQiOiBbeyJSZXNvdXJjZSI6IiovdS9laTJIMTN2UjFjVjQ2TTNOdDU2Q3pkLyoiLCJDb25kaXRpb24iOnsiRGF0ZUxlc3NUaGFuIjp7IkFXUzpFcG9jaFRpbWUiOjE2NjE1NTg2NDB9fX1df"
    ]
}

Parameters

Parameter	Type	Tags	Description
race	array of strings	Query, Optional	Racial Options: ["asian", "white", "middle eastern", "indian", "latino", "black"]
rating	float	Query, Optional	0.0 - 5.0
images	array of strings	Body, Requred	String can be the image URL or Base64 encoded image. Limit of up to 10 images per request.

Response

{
    "match": true, 
    "race": "white", 
    "rating": 2.5576894283294678
}

Example Usage

flask run

Bot

Signs in to Tinder, grabs potential match profile pictures, sends them to flask server, and once the server responds the bot will either swipe left or right.

Required Environment Variables

• FB_USERNAME
• FB_PASSWORD

Command-Line Flags

Flags	Type	Description
remote	bool	Open Chrome browser directly on device or start remote server to running chromedriver
host	string	Host runnning API
port	string	Port to access API
use_api	bool	Use API or interact with browser
human_login	bool	Automated or human interaction with browser
pref_race	array	Preferred race filter
min_rating	float	Minimum match rating
like_count	int	Amount of likes before exiting browser

Example Usage

python3 tinder_bot.py --min_rating 2.55