Physics LLM 🪀

A fine-tuned LFM2-350M predicts 2D rigid-body physics frame-by-frame, as text — no physics engine. Pick a preset, drag the balls around or drop new ones on the canvas, then watch the model roll it forward live. The model sees as much of the prior trajectory as fits its 8192-token context.

Six scenarios (pong, bowling, ramp_roll, angry_birds, hourglass, newtons_cradle) were never seen in training.