---
base_model:
- nvidia/Eagle2-2B
---


# Pretrained VLM Model and Adapter for InstructVLA

**Eagle2-2B:** Uses the same pretrained weights as [https://huggingface.co/nvidia/Eagle2-2B](https://huggingface.co/nvidia/Eagle2-2B), with a modified `modeling_eagle_chat.py` for bug fixes.

**empty_language_adapter:** An empty LoRA adapter for Eagle2-2B, used in Stage-2 training for InstructVLA.

Please move these two folders to `path_to_project_root/ckpt`.