AIRoA-0 (A-0) is a bimanual manipulation system learned from 16,151 hours of handheld demonstrations, with no robot data collected at any stage. The demonstrations are recorded with YUBI, our open-source handheld gripper, which lets people record manipulation data anywhere with no robot present. From a single instruction, A-0 sorts more than a dozen small parts into their numbered cells and changes the mop pads and brush of a robot vacuum; among other tasks, it unzips a pencil case and folds a towel. It performs these tasks using only a camera on each wrist, with no head or scene camera, at a site where none of its training data was recorded. Across the ten tasks of our evaluation, it averages a success rate of 78%, against 27% for the strongest of four other vision-language-action models post-trained on the same task demonstrations. We collect no teleoperation data and use no reinforcement learning or on-robot corrections, and the robot appears only at test time. The system learns which regions of each wrist view matter for the task, predicts the stage the task has reached, and decodes actions as a continuous stream. A-0 is embodiment-agnostic: it runs on robots that mount the YUBI gripper, such as a bimanual Franka, TX-G2, OpenArm and UR. We are releasing the dataset, to our knowledge the largest open dataset of handheld-gripper demonstrations with task and subtask instructions, together with the pre-trained weights and the post-training recipe. Videos of A-0 on every task are on the project page, https://a-0.airoa.io. Anyone can build the gripper, record a task of their own and post-train the model on it in under four hours on a single node of eight H200 GPUs.
Figures & tables
Figure 1: A-0 in overview. Handheld demonstrations, seen here through their wrist cameras, are the only data we collect. The four parts of the system turn them into actions for robots that mount the same gripper and cameras; the photographs show towel folding on the Franka, the pencil case on TX-G2, colour sorting on OpenArm and the orange pyramid on UR. The chart gives each model’s success rate, averaged over the ten tasks of Section 2.1 .
Figure 2: The ten evaluation tasks, each in a frame from an evaluation run of A-0 .
Data (h)
A-0
π 0.5
XR-1
GR00T N1.7
MolmoAct2
Towel folding
3.3
73
22
21
0
8
Letter blocks
3.4
84
13
0
24
0
Orange pyramid
3.4
90
2
52
13
28
Parts kitting
4.0
83
30
13
7
12
Colour sorting
4.1
97
45
39
20
19
Robot vacuum
5.3
73
1
14
1
3
Table 1: The ten evaluation tasks: the hours of handheld demonstrations of each task that post-training uses, and the success rate (%), each an average over 15 trials. A-0 uses the one recipe of Table 4 on every task; the other four models use their best setting for each task. The best entry for each task is in bold.
Figure 3: Two minutes of one autonomous run of the robot-vacuum task, with the time on the robot since (a). (a) The gripper takes hold of the first used mop pad, and (b) the second comes off. (c) A new pad is carried from the box and (d) pressed onto its mount; (e) both new pads are on, and (f) the brush is pulled out.
Figure 4: The bins moved during autonomous colour sorting, with the time on the robot since (a). (a) An arm carries a green part to the green bin; the bins are, from left to right, yellow, green and red. (b) A person pushes the bins to the left, and the red bin is now under the gripper. (c) The arm moves over to the green bin and (d) drops the part into it.
Figure 5: A-0 on TX-G2, OpenArm and UR, each task in a frame from a run on that robot.
Figure 6: A-0 as one system. Handheld demonstrations pass through data processing into pre-training; post-training adapts the pre-trained model to the tasks it will perform; the inference harness runs it on the robot.
Demonstrations
16,151 hours, bimanual, recorded with YUBI
Subtask segments
10.65 million, 5.5 seconds on average
Time steps
1.74 billion at 30 Hz
Views
two wrist cameras
Proprioception
pose and opening of each gripper
Language
a task instruction and a subtask instruction for every segment
Table 2: The corpus.
Figure 7: The A-0 model. Pre-training adds the visual budget gate and subtask prediction to the π0.5 backbone. Post-training adds the stage head, whose prediction conditions the action expert as text, and the proprioceptive memory.
Figure 8: Where the visual budget gate puts its weight, on 25 subtasks. Each tile shows the left and right wrist views the model receives. The contours trace the smoothed keep probability at three levels, and the view fades where the gate gives it less weight.
Figure 9: Action loss on held-out demonstrations for models pre-trained on a share of the data. Left, during pre-training, with the task instruction as prompt (the subtask instruction gives the same picture): every run has the same model, batch size and schedule, and the 100% run is the pre-training of A-0 . Right, during post-training, averaged over 16 task sets: each run starts from its checkpoint at 100,000 pre-training steps, and post-training data and settings are the same for all.
Figure 10: Post-training action loss on held-out demonstrations against the amount of pre-training data, on logarithmic axes. Each point is the 16-set average in the right panel of Figure 9 over its last 1,000 updates; the dashed line is a least-squares power-law fit ( R2=0.98 ).
Task
Place each specified YUBI part into its corresponding numbered cell.
Stage
Subtasks
Place the fingers, flaps and gears in cells 1, 2 and 3.
Pick the finger part and place it into cell 1. Pick the flap part and place it into cell 2. Pick the gear part and place it into cell 3.
Place the shafts, gear boxes and Quest holder in cells 4, 5 and 6.
Pick the shaft part and place it into cell 4. Pick the gear box part and place it into cell 5. Pick the Quest holder part and place it into cell 6.
Place the camera brackets and grip in cells 7 and 8.
Pick the camera bracket part and place it into cell 7. Pick the grip part and place it into cell 8.
Table 3: The parts-kitting task at three levels of language. The task instruction is all the robot is given; the stage is predicted by the model; the subtask instructions label the demonstrations.
Initialisation
pre-trained weights; stage head and memory from scratch
Action chunk
20 anchored actions at 20 Hz
Prompt
task instruction
Visual evidence budget
ρ=0.9 , γ : 0.1→0.02 , τ=0.805 , λb=0.1
Loss weights
stage λc=0.1 , subtask λs=0.01
Proprioceptive memory
ten past states, four memory tokens
Updates / batch size
20,000 / 64
Table 4: Post-training recipe.
Figure 11: Streaming decoding with two slots. Left: the buffer before and after one query. Right: the noise level of one chunk on its way through the buffer, with a uniform schedule and with the non-uniform schedule used in A-0 .
Appendix figures & tables1 asset
Supplementary material from the paper’s appendix.
Appendix
Figure 12: The loss in the right panel of Figure 9 for each of the 16 task sets.