EDA-env: A Reinforcement Learning Environment for RTL-to-GDS Design with Independent Scoring

Task generator, independent scorer, 103 qualified RTL-to-GDS task packages on ASAP7, Nangate45 and Sky130HD, the scorer regression suite (143 probes), and the recorded Kimi-k3 episodes (patches, trajectories, sealed measurements) reported in the ICCAD 2026 paper "EDA-env: A Reinforcement Learning En... View More