The sidecar receives VLABench's front, second, and wrist views at 480x480 and the raw 7-D relative EE state. XR-1 predicts ten 7-D deltas. OpenRAL integrates those deltas into the absolute targets consumed by VLABench and executes five before replanning. The gripper scalar uses Xiaomi's 0.2 threshold and maps to OpenRAL's normalized closed/open commands.

Camera order is load-bearing and matches Xiaomi's evaluator: camera1=front(raw index 2), camera2=base/second(raw index 0), camera3=wrist(raw index 3). Frames are not flipped (flip_180: false).

Three RGB streams at 480x480 and VLABench's 7-D EE/gripper state.

Provision the VLABench asset bundle documented by the scene backend, then run the command above. This package ships no score until that real simulator run has completed.

The default sidecar quantizes the pinned BF16 source at load. Export once to avoid that repeated pack step:

The local unpublished package is then outputs/run_artifacts/xr1-vlabench-nf4-rskill. Its packed output was verified bit-identical to runtime NF4 on the same seeded input (MAE=0, max error=0). On the reference host, policy initialization dropped from about 18.0 s to 10.1 s.

NF4 was loaded and executed on an RTX 4070 Laptop 8 GB against the real VLABench select_fruit environment. The sidecar used 3.66 GiB process VRAM; the first synthetic chunk took 1.35 s and the warmed real-scene chunk took 0.81 s. One finite (10, 7) chunk was integrated and applied successfully. This validates execution, not task success or Xiaomi's benchmark score.

The wrapper and upstream code/weights are Apache-2.0. Loading requires OPENRAL_ALLOW_REMOTE_CODE=1 for the checkpoint's custom Transformers code.