Update model card

README.md

@@ -0,0 +1,38 @@
+---
+library_name: lerobot
+tags:
+- molmoact2
+- robotics
+- lerobot
+- vla
+
+base_model: allenai/MolmoAct2-SO100_101
+---
+
+# molmoact2_tool_insert_20k
+
+Fine-tuned [MolmoAct2](https://huggingface.co/allenai/MolmoAct2) (action-expert-only) for `tool_insert` on SO101 data.
+
+| | |
+|---|---|
+| **Policy** | MolmoAct2 (`policy.type=molmoact2`) |
+| **Init checkpoint** | [allenai/MolmoAct2-SO100_101](https://huggingface.co/allenai/MolmoAct2-SO100_101) |
+| **Dataset** | [villekuosmanen/armnetbench_tool_insert](https://huggingface.co/datasets/villekuosmanen/armnetbench_tool_insert) |
+| **Task** | `tool_insert` |
+| **Action dim** | 6 (single-arm) |
+| **Cameras** | `top`, `wrist`, `front` |
+| **Training** | 20k steps, batch 32 (8/GPU × 4 DDP), 4× RTX 5090 on Vast.ai |
+| **Prior HF repo** | [pravsels/molmoact2_tool_insert](https://huggingface.co/pravsels/molmoact2_tool_insert) |
+| **W&B project** | [molmoact2_tool_insert](https://wandb.ai/pravsels/molmoact2_tool_insert) |
+| **W&B run** | [vsh6rneo](https://wandb.ai/pravsels/molmoact2_tool_insert/runs/vsh6rneo) |
+
+## Checkpoints
+
+The final checkpoint (local step `020000`, 20k total steps) lives at the **repository root** for direct loading.
+
+## Usage
+
+```python
+from lerobot.policies.molmoact2.modeling_molmoact2 import MolmoAct2Policy
+policy = MolmoAct2Policy.from_pretrained("pravsels/molmoact2_tool_insert_20k")
+```