Tesseracted LabsTry the chat ↗
KOLIBRI-1 · RECORDED DEMO

Less talk. More Breakout.

Recorded gameplay. The live experiment has ended. This clip replays an earlier run.

Probabilities, not generated text.

Kolibri-1 reads the ball and paddle coordinates, scores four actions, and follows a paddle-tracking policy supplied in the prompt.

We used vLLM’s classification path to score action labels, without generating a text response. Screenshots are for spectators; the model receives structured state.

How the experiment worked

The scoring head reused selected rows of the model’s existing output weights. It was not a newly trained game-playing classifier. Scores are conditional on the supplied options and are not calibrated confidence. The model and controller both contribute to the gameplay shown.