screenquest: a local vision game agent for Apple silicon

Combines screenshot perception, Laya decisions on Core ML and Qwen planning to play, click loot and review its own evidence, entirely locally.

Open on GitHub

Preview of screenquest: a local vision game agent for Apple silicon

Launch: Laya
Use case: Games
Format: Build
Source: github.com

Daniel Tremer

Managing Director at control-f GmbH | Building AI Agents, Automation & Knowledge Graphs | Data Science & ML | Berlin

See all

雨夹雪❄️

@mizorewww · Sep 20

The port announcement: Laya on MLX, under 1 GB RAM, choosing the snake’s next move 60 times a second on a local M3 Max.

atomic.chat

@atomic_chat_hq · Sep 21

Atomic.chat ran Laya on a 16 GB MacBook Air against Jev over the API on a Grok-built Tetris. Local won on speed: 11x faster decisions.

Viraj Bhartiya

@heyxviraj · Sep 21

Viraj Bhartiya wired both models into the Chrome T-Rex game: Laya local on MLX, Jev over the API, same track and physics, side-by-side latency. Code in virajbhartiya/laya-vs-jev.

mizorewww

@mizorewww · Sep 19

Validated Core ML ports of all Laya checkpoints. One short multilingual decision in 4.98 ms P50 on M3 Max with ANE FP16, at 2.78x better energy per decision than MLX FP16.

Open on GitHub

screenquest: a local vision game agent for Apple silicon

More built with Laya