Update README.md

README.md

@@ -1,5 +1,69 @@
----
-license: other
-license_name: other
-license_link: LICENSE
----
+---
+license: other
+license_name: other
+license_link: LICENSE
+---
+# MolmoBot-SPOC-RBY1Articulated
+
+[[Paper](https://arxiv.org/pdf/2603.16861)] [[Project Website](https://allenai.github.io/MolmoBot)] [[Code](https://github.com/allenai/MolmoBot/tree/main/MolmoBot-SPOC)] [[Data](https://huggingface.co/datasets/allenai/molmobot-data)]
+
+MolmoBot-SPOC-RBY1Articulated is the MolmoBot-SPOC model trained on simulation data on the RB-Y1 platform, **without any real robot data**. See [here](https://github.com/allenai/MolmoBot/tree/main/MolmoBot-SPOC) for usage instructions.
+
+## Quickstart
+
+
+```python
+import numpy as np
+from huggingface_hub import snapshot_download
+from molmobot_spoc.eval.config.rby1_eval_config import RBY1EvalBaseConfig
+from molmobot_spoc.eval.config.spoc_policy_configs import SPOCRBY1ArticulatedManipPolicyConfig
+from molmobot_spoc.eval.spoc_policy import SPOCModelPolicy
+
+
+ckpt_dir = snapshot_download("allenai/MolmoBot-SPOC-RBY1Articulated")
+policy_config = SPOCRBY1ArticulatedManipPolicyConfig(checkpoint_dir=ckpt_dir)
+config = RBY1EvalBaseConfig(
+    policy_config=policy_config, task_type="open"
+)  # Try with "door_open" too!
+
+policy = SPOCModelPolicy(config, config.task_type)
+
+obs = {
+    "head_camera": np.zeros((576, 768, 3), dtype=np.uint8),
+    "wrist_camera_r": np.zeros((576, 768, 3), dtype=np.uint8),
+    "wrist_camera_l": np.zeros((576, 768, 3), dtype=np.uint8),
+    "qpos": {
+        "left_arm": np.zeros((7,), dtype=np.float32),
+        "right_arm": np.zeros((7,), dtype=np.float32),
+        "torso": np.zeros((6,), dtype=np.float32),
+        "left_gripper": np.zeros((2,), dtype=np.float32),
+        "right_gripper": np.zeros((2,), dtype=np.float32),
+    },
+    "goal": "open the drawer",
+    "object_image_points": {
+        "pickup_obj": {  # NOTE: Use "pickup_obj" for open tasks, "door_handle" for door_open tasks.
+            "head_camera": {
+                "points": [[0.45, 0.52]]  # list of [x, y] candidates; one is sampled at runtime
+            }
+        }
+    }
+}
+# NOTE: get_action saves an internal buffer with the chunk and returns actions 1 by 1. To get the whole chunk, use model_output_to_action
+# NOTE: the policy internally saves the first frame to ground the image point with
+action = policy.get_action(obs)
+print(action)
+```
+
+## BibTeX
+
+```
+@misc{deshpande2026molmobot,
+      title={MolmoB0T: Large-Scale Simulation Enables Zero-Shot Manipulation},
+      author={Abhay Deshpande and Maya Guru and Rose Hendrix and Snehal Jauhri and Ainaz Eftekhar and Rohun Tripathi and Max Argus and Jordi Salvador and Haoquan Fang and Matthew Wallingford and Wilbert Pumacay and Yejin Kim and Quinn Pfeifer and Ying-Chun Lee and Piper Wolters and Omar Rayyan and Mingtong Zhang and Jiafei Duan and Karen Farley and Winson Han and Eli Vanderbilt and Dieter Fox and Ali Farhadi and Georgia Chalvatzaki and Dhruv Shah and Ranjay Krishna},
+      year={2026},
+      eprint={2603.16861},
+      archivePrefix={arXiv},
+      primaryClass={cs.RO},
+      url={https://arxiv.org/abs/2603.16861},
+}
+```