| library_name: lerobot | |
| tags: | |
| - smolvla | |
| - robotics | |
| - lerobot | |
| base_model: lerobot/smolvla_base | |
| # smolvla_tool_insert | |
| Fine-tuned [SmolVLA](https://huggingface.co/lerobot/smolvla_base) on SO101 joint-space data. | |
| | | | | |
| |---|---| | |
| | **Base model** | [lerobot/smolvla_base](https://huggingface.co/lerobot/smolvla_base) | | |
| | **Dataset** | [villekuosmanen/armnetbench_tool_insert](https://huggingface.co/datasets/villekuosmanen/armnetbench_tool_insert) | | |
| | **Checkpoint step** | 10,000 | | |
| | **Training** | Isambard-AI GH200, batch 128 | | |
| | **W&B project** | [smolvla_tool_insert](https://wandb.ai/pravsels/smolvla_tool_insert) | | |
| | **W&B run** | [bt0cbydm](https://wandb.ai/pravsels/smolvla_tool_insert/runs/bt0cbydm) | | |
| ## Usage | |
| ```python | |
| from lerobot.policies.smolvla.modeling_smolvla import SmolVLAPolicy | |
| policy = SmolVLAPolicy.from_pretrained("pravsels/smolvla_tool_insert") | |
| ``` | |