Small decisions. Interesting possibilities.Submit contentSubmit

Chrome Dino with live decision scores

@typesafeai jev playing chrome dino! the model is incredibly fast! - avg response time: ~280ms - spent 116M tokens - cost? just $4.65. - the input here is a snapshot of the chrome runner - jev decides whether to jump, run or duck with a confidence score repo link if you want to try it out! 👇🏻

Added to Jevfast

How it uses Jev

Jev receives snapshots of the Chrome runner and chooses whether the game should jump, run, or duck, with a confidence score. The post reports an average response time but does not explain the runner integration.

App workflow

  1. Capture runner state

    The input is a snapshot of the Chrome runner.

  2. Choose a game action

    Jev selects jump, run, or duck and returns a confidence score.

  3. Continue the game loop

    The chosen action is used to play Chrome Dino; the post reports an average response time of about 280 ms.

Why it is interesting

It turns a compact screenshot-based game state into a small action set that can be applied continuously during play.