BEST-POLICY RACE

The fastest published policy from each model

Four policies start together on the same 100-metre clock.

0.00s

Lane 1 · DeepSeek V4 Flash Vision ExpDNF 60.00s0.0 m

Lane 2 · GLM‑5.3‑Flash13.06s0.0 m

Lane 3 · GPT‑5.6 Sol13.12s0.0 m

Lane 4 · Claude Opus 5DQ 0.95s0.0 m

Best replayable policy per model from the current published cohort.Whole body must stay between the ±0.61 m vertical lane planes; scoring stops at the verifier eventDrag to orbit, scroll to zoom, space to pause

The replay uses the verifier-authored pose capture for every runner.

Best published policies racing together in lanes one through four.

The fastest published policy from each model