Laya-MLX is an open source classification system similar to

Built Laya-MLX, an open source on-device classifier modeled on Jev, running at 60 decisions per second in 1GB of memory.
The original post
Laya-MLX is an open source classification system similar to Jev but 50x faster, running locally on device with a maximum of 1GB memory and making decisions at 60 per second, fast enough to play Snake on an M3 Max at full speed. Repo here: https:// github.com/mizorewww/laya -mlx …
Engagement when collected
Numbers are a snapshot taken from X when the case was added to the library (schema v1, collected 2026-09-19); they will not match today.
Where this case fits
Filed under games & playable demos, coding & developer tools, classification & data extraction. In the pattern Jev is built for, the model answers a bounded question per step — and ordinary code acts on the answer, because the answer is already a value rather than a paragraph. Other posts in the same family are on the games & playable demos page.
Related Jev cases
A fully autonomous real-time trading bot built with Jev in
Finance
This is what happens when a designer gets access to Jev.
Jev combined with Grok Bot makes the agent setup cheaper
Agents
The gains aren’t free: Jev can't generate text Comparing
Dev toolsClassification
Keep browsing: all 1173 Jev cases · more from @RoundtableSpace · builders · what Jev is
Last updated: 2026-09-22 · sources & corrections · every card links to its author's original post