DEMO / RECORDING
Overview & result
Nailthy shared a Drape virtual try-on experiment. The user speaks; Jev reads the transcript and current outfit, chooses items from a closet and changes the outfit in the demo. The author reports about $0.0011 and 620ms per decision.
From the original post
jev is insane 🫣
it makes realtime virtual try-on hauls possible.
built this experiment for Drape with @typesafeai
> i talk
> jev reads transcript + what i'm wearing
> picks from my closet
> changes my outfit in realtime
cost: $0.0011 per decision
time: ~620ms per decision
imagine getting ready like this:
How Jev fits in the loop
- Capture the spoken request and current outfit information.
- Use Jev to choose items from the closet.
- Show the updated look in the virtual try-on demo.
How to reproduce
- Watch the original demonstration on X.
- The current post does not link a public implementation or reproducible benchmark.
Why this build matters
The demo connects a voice request to a rapid outfit-selection decision.
Reported performance
Reported by authorCost: $0.0011 / decision (author claim)
Latency: ~620ms / decision (author claim)
The author reports $0.0011 and roughly 620ms per decision; the post does not identify a rendering provider or break down the latency.
Limitations
- The reported cost and time are author-provided figures for the demo.