ASPIRE exploration research archive

This repository preserves selected self-trained artifacts from an exploratory robotics project. It does not redistribute ASPIRE's complete learned skill library or third-party pretrained models.

Available archive

2026-10-27: four rejected diagnostic world-model checkpoints. All four failed state-prediction validation. They are retained as negative-result evidence, not recommended as working dynamics models or robot policies. See the archive's manifest and validation summary for exact hashes, source and limitations.

Downloads last month

-

Downloads are not tracked for this model. How to track
Video Preview
loading