Under the hood

How the AI judges your drawings

Rating a 60-second finger doodle fairly is harder than it sounds. Here's exactly how Sketchmate turns your scribble into a score, and why you can trust it.

How the Sketchmate AI judge scores a drawing A drawing goes through two passes in parallel. The blind pass guesses what it shows without knowing the answer. The rubric pass scores subject, details and effort against the secret. Both combine into a 0 to 100 score with an explanation. Your drawing 1 · Blind guess “What is this?” without being told the answer Top-3 guesses → recognition score (30%) 2 · Rubric vs. the secret Subject 60% · Details 25% · Effort 15% Blank = 0 · written answer capped at 20 82 / 100 ✓ AI guessed it blind + bushy tail − orange colour plus a one-line roast
Two passes run in parallel. The final score is computed from the numbers, not made up by the model, so it's consistent and explainable.

Pass 1: the blind guess

The first thing we ask a vision model is simply “what is this?”, without telling it the answer. Vision models told the answer tend to “see” it in almost anything. A blind guess avoids that, and it's the same test your friends are playing: would a stranger recognise it?

The model returns its top three guesses. If the secret shows up, you earn recognition points: full marks for the first guess, 75% for the second, 55% for the third. Close answers count too. “Tree” earns 70% for “Pine Tree”.

Pass 2: the rubric

In parallel, a second pass knows the secret and the describer's clue. It scores three things from 0 to 10:

CriterionWeightWhat it measures
Subject60%Does the drawing show the secret thing?
Details25%Did you capture the traits from the clue: shapes, colours, parts, setting?
Effort15%Is it complete, with a clear composition? One squiggle is not a masterpiece.

It also names what matched (“bushy tail”) and what's missing (“orange colour”). Those phrases appear on your results card.

The score is calculated, not invented

The model never picks the headline number. Sketchmate computes it from the rubric (70%) and the blind recognition (30%), with a few fairness rules:

82Right subject, most clue details, AI guessed it blind.
18Lovely rocket, but the secret was a cupcake.
20Wrote “FOX” in huge letters. Nice try.

Built to be fair

The judge runs on Meta's Llama 4 Scout vision model on Cloudflare Workers AI, with Llama 3.2 Vision as an automatic fallback.

Scoring points in the game: 85+ is a Masterpiece (10 pts), 60+ Not Bad (6), 35+ Trying Hard (3), anything else is Beautiful Chaos (1). The describer earns a bonus when the table's average is above 50 or 70.

Try to fool it

Grab 2–8 friends and see who the AI can read. Free on iPhone and Android.

Download on theApp Store GET IT ONGoogle Play