rubiks-reasoning-latency-bench

Benchmarking fast System-1 decision models (Jev, GPT-6 Luna, Clef) against a frontier LLM (Claude Opus 5.5) on Rubik’s Cube solving via OpenRouter. Measures per-move latency, solve rate, and optimality gap vs. BFS-optimal solutions.

Stars

0

Forks

0

Open issues

0

Last push

today

Star history starts building from the first daily refresh.