Robotics
Transformers

OpenVLA-OFT+ Domain-Adapted Checkpoints

This repository stores independently trained OpenVLA-OFT+ adapters and their action-head and proprio-projector components.

No production or submission candidate is published on main yet. Bounded environment-validation checkpoints are kept on branches whose names start with smoke-colab-20step.

Provenance

  • Base model: argo11/parc2026-openvla-oft-plus-base
  • Base revision: 0d923bf7af68b7eac9290cdd65f597cad7b48596
  • Training dataset: argo11/openvla-oft-plus-robustness-recovery
  • Dataset revision: c3807717b10ca28c56630508262af7163ca59d45
  • OpenVLA-OFT code revision: e4287e94541f459edc4feabc4e181f537cd569a8

The dataset is published with license: other; its upstream-specific terms are not fully resolved. Public availability does not by itself grant unrestricted reuse. See the dataset card and third-party notices before using these weights. The pinned base Model Card declares the base weights under MIT. Because the training dataset's aggregate license remains unresolved, this repository also uses license: other; the base-model MIT declaration does not resolve data or third-party asset rights for the derived checkpoint.

No related online W&B run is claimed for the Colab smoke validation. Training uses W&B offline mode unless a later release documents another provenance path.

Downloads last month

-

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

Model tree for argo11/openvla-oft-plus-domain-adapted

Finetuned
(1)
this model

Dataset used to train argo11/openvla-oft-plus-domain-adapted