An ESP32 and a Local Vision Model Drive My iPhone

Automated excerpt

The board exposes two HID interfaces via TinyUSB: relative mouse + ASCII keyboard. Step 3:... " The actor (smaller, faster) only executes step 2 given the current screen. This is how Yuan's app runs: decode → vision model → action — a planner and an actor: tap, swipe, draw, type, press keys, go home, finish On one GPU, run the planner every 3 steps, not every step.

Selected automatically from source text; not independently written or fact-checked. Read the original for full context.