LingBot-VLA 2.0 Base Model for LeRobot

lerobot/lingbot_vla_v2_base

Summary

LingBot-VLA 2.0 base model uses Qwen3-VL-4B with sparse-MoE action expert. Pre-trained for fine-tuning on new robots. Licensed under Apache-2.0.

Source: LeRobot — Model repository updates · Read original article ↗

Article text · Original source · English

On this page

LingBot-VLA 2.0 (base)

LingBot-VLA 2.0 converted for LeRobot (policy.type=lingbot_vla_v2): a Qwen3-VL-4B backbone with a sparse-MoE Qwen2 action expert, trained with flow matching.

Pre-trained base checkpoint, the starting point for fine-tuning on a new robot.

The weights are the upstream robbyant/lingbot-vla-v2-6b weights, unchanged (fp32, every tensor checked against upstream), with a LeRobot config and processors.

Usage

lerobot-train \
  --policy.path=lerobot/lingbot_vla_v2_base \
  --dataset.repo_id=<your_dataset> \
  --policy.state_slots=... --policy.action_slots=... \
  --policy.repo_id=<your_repo_id>

See the docs for the slot mapping of your robot.

License

Apache-2.0, as the upstream release. Fine-tuning with the dual-query distillation (on by default) downloads frozen teachers under their own licenses, including DINO-Video weights under the DINOv3 License. Inference does not use them.

Citation

@article{lingbotvla2,
  title={From Foundation to Application: Improving VLA Models in Practice},
  author={Wei Wu and others},
  journal={arXiv preprint arXiv:2607.06403},
  year={2026}
}