Features
Blind pairwise voting on AI-generated designs
Crowdsourced leaderboards updated in real-time
Tournament format with four anonymous models per session
Bradley-Terry ranking algorithm
190+ countries participating
Multiple design categories: Website, Game Dev, 3D, Video, Slides, Text-to-Speech, Mobile Apps, and more
Hidden model identity to prevent brand bias
Builder category for one-shot prompt comparison
API for automated voting and data access
Router feature to direct prompts to best model
Community evaluation via Discord, Twitter, Reddit
Changelog tracking model additions/removals
Voice conversation via Text-to-Speech category
Vision/image understanding in Image and UI Component categories
Image generation compared across models in Image category
160+ beauty marker facial analysis
Personalized non-surgical glow-up protocol
Visual projection of best-looking self
Research-backed recommendations with citations
Tailored to ethnic background and demographics
Considers lifestyle factors (diet, stress, sleep)
Adjusts for natural aging patterns
Facial dimorphism analysis
Detailed ear shape and projection analysis
Refined eye scleral and limbal ring measurements
Sharper jaw geometry angle and inclination analysis
Lip texture and health analysis
Refined eyebrow analysis with Westmore boundaries
Face shape analysis section
