new FACET-0 post-training model upload

README.md

@@ -1,3 +1,80 @@
 ---
 license: apache-2.0
+pipeline_tag: robotics
+tags:
+- robotics
+- vision-language-action
+- openpi
 ---
+
+# Facet-0
+
+Model weights for [Facet-0: A Robotic Foundation Model for Contact-Rich Precise Manipulation](https://arxiv.org/abs/2609.01596).
+
+## `facet0-post-training`
+
+Policy produced by Facet-0 post-training (value-guided policy refinement). A PaliGemma vision-language backbone is paired with a flow-matching action expert. The kinematic state and a causal wrench history enter only the action expert, which generates each 50-step action chunk together with the wrist wrench it is expected to induce. Only the action is executed. During refinement the vision-language backbone is frozen.
+
+| File | Content |
+|---|---|
+| `params/` | weights (orbax checkpoint; frozen backbone stored in bfloat16) |
+| `assets/norm_stats.json` | normalization statistics used at inference |
+| `config.json` | input / output description |
+
+### Inputs
+
+| Key | Shape | Content |
+|---|---|---|
+| `observation/state` | float32 `[13]` | end-effector pose (6: x, y, z, rotation vector), gripper (1), wrist wrench (6) |
+| `observation/wrench_history` | float32 `[10, 6]` | latest 10 measured wrist wrenches, oldest first |
+| `observation/image` | uint8 `[H, W, 3]` | third-person view |
+| `observation/wrist_image` | uint8 `[H, W, 3]` | wrist camera |
+| `observation/extra_view_image` | uint8 `[H, W, 3]` | second third-person view |
+| `prompt` | string | natural-language sub-task instruction |
+
+### Outputs
+
+| Key | Shape | Content |
+|---|---|---|
+| `actions` | `[50, 7]` | absolute end-effector pose + gripper; the only commanded output |
+| `wrench` | `[50, 6]` | anticipated wrist wrench (N, N m); predictive, never commanded |
+
+### Usage
+
+The checkpoint loads with the Facet-0 post-training code, which builds on [openpi](https://github.com/Physical-Intelligence/openpi).
+
+```python
+import facet0.policy
+
+policy = facet0.policy.load("facet0-post-training")      # params/ + assets/norm_stats.json
+history = facet0.policy.WrenchHistory()                   # latest 10 wrenches
+history.push(state[7:13])
+out = policy.infer({
+    "observation/state": state,
+    "observation/wrench_history": history.window(),
+    "observation/image": view1,
+    "observation/wrist_image": hand,
+    "observation/extra_view_image": view2,
+    "prompt": "press down the RAM until it locks into place",
+})
+out["actions"]   # [50, 7]: send to the controller
+out["wrench"]    # [50, 6]: for logging
+```
+
+`guidance_scale` trades between the unconditional branch (0.0) and the label-conditioned branch (1.0, default).
+
+## Citation
+
+```bibtex
+@article{deng2026facet0,
+  title   = {Facet-0: A Robotic Foundation Model for Contact-Rich Precise Manipulation},
+  author  = {Deng, Haoyuan and Liu, Haichao and Guo, Wenkai and Ling, Yuan and Yang, Zaijia and
+             Xue, Yuanjiang and Sun, Haosheng and Wang, Liangzi and Wang, Ziwei},
+  journal = {arXiv preprint arXiv:2609.01596},
+  year    = {2026}
+}
+```
+
+## License
+
+Apache-2.0. The model builds on openpi (Apache-2.0).

facet0-post-training/config.json

@@ -1,11 +1,19 @@
 {
-  "model_name": "Post-training model for FACET-0, which includes wrench-aware value estimation and ensembling, with wrench-conditioned RL goal for evolving on real-world rollouts",
+  "model_name": "Post-training model for FACET-0, which includes wrench-aware value estimation, with wrench-conditioned RL goal for evolving on real-world rollouts",
   "model_type": "vision-language-action",
   "license": "apache-2.0",
   "library": "openpi",
   "observation": {
     "state_dim": 13,
-    "state_layout": "end-effector pose (7) + six-axis wrench (6)",
+    "state_layout": "end-effector pose (6: x, y, z, rotation vector) + gripper (1) + six-axis wrist wrench (6: force N, torque N m)",
+    "wrench_history": {
+      "key": "observation/wrench_history",
+      "shape": [
+        10,
+        6
+      ],
+      "layout": "latest 10 measured wrist wrenches (force N, torque N m), oldest first"
+    },
     "images": [
       "observation/image",
       "observation/wrist_image",
@@ -15,11 +23,7 @@
   },
   "action": {
     "horizon": 50,
-    "executed_channels": "absolute end-effector pose + gripper",
-    "gripper_convention": {
-      "0": "closed",
-      "1": "open"
-    }
+    "executed_channels": "absolute end-effector pose + gripper"
   },
   "inference": {
     "guidance_scale_default": 1.0,