Where does it go?

Drag each thing into a bin. A decision model plays against you, scored against the City of Toronto's own published waste rules. Every item is pre-scored, so this page runs offline with no server and no API key.

Jev accuracy 71.2% obvious keyword rule 39.5% 1,840 items scored whole A to Z in 58s for $0.065 at 95%+ confidence 94.4%

Drag it to a bin, or click a bin, or use the keyboard: Tab to a bin then Enter.

You 0 Jev 0 Round 1/20

show only where Jev disagrees with the City

item	City says	Jev said	conf.

These objects appear twice in the City's list with different qualifiers and land in different bins. Change one word and the answer changes. A keyword rule got 0 of 38 objects fully right, because a keyword rule has no idea what a qualifier is. Jev got 23 of 38.

The blind spot: it will not tell you no

When the City's answer is that it does not accept something at all, Jev finds that answer only 8.7% of the time (103 items). It sends them to a drop-off depot instead. It would rather hand you an action than deliver a refusal. The same bias showed up in a separate health project, where it barely used its lowest category.

What it gets wrong most often

City says	Jev said	count

Counterintuitive rules it misses

Toronto says an electric toothbrush with removable batteries goes in the Garbage Bin, and with non-removable batteries it becomes Household Hazardous Waste. Jev sent both to electronics drop-off. Small batteries are also one of the most common causes of fires in collection trucks, so this is exactly the kind of rule worth getting right.

So is it useful?

At 71.2% overall, no: one item in three would be wrong. But at 0.95 confidence and above it is 94.4%, and below 0.5 it is 43.4%. Read as a suggestion that a person confirms, with the City's rule shown alongside, it is a genuine shortcut. Read as an oracle, it is a contamination risk.