MolmoAct 2 Evaluation gallery
IN PROGRESS · PROVISIONAL Only fully completed, validated task shards are shown. Campaign coverage and aggregate rates can change.
Supplemental Task 012 Validated task

put the black bowl at the back on the plate

Completed task success rate
64.0%
32 of 50
Control horizon
400
Maximum steps
Camera videos
100
Two per rollout
Evaluation time
9m 33s
Task process wall time

Canonical catalog

Task definition

Model instruction
put the black bowl at the back on the plate
BDDL language
put the black bowl at the back on the plate
Problem
libero_kitchen_tabletop_manipulation
Objects
akita_black_bowl_1, akita_black_bowl_2, akita_black_bowl_3, plate_1
Fixtures
kitchen_table, wooden_cabinet_1
Initial-state predicates
5
Fixed init states
50
Exact initial-state predicates 
(on akita_black_bowl_1 kitchen_table_akita_black_bowl_front_init_region)
(on akita_black_bowl_2 kitchen_table_akita_black_bowl_middle_init_region)
(on akita_black_bowl_3 kitchen_table_akita_black_bowl_back_init_region)
(on plate_1 kitchen_table_plate_init_region)
(on wooden_cabinet_1 kitchen_table_wooden_cabinet_init_region)

Open exact BDDL definition

Simulator truth

Completion criteria

LIBERO checks these BDDL goal predicates after every action. All conditions must be true simultaneously in the same simulator step for the rollout to succeed.

  1. akita black bowl 3 is on plate 1

    (on akita_black_bowl_3 plate_1)
Exact parsed BDDL goal state 
(on akita_black_bowl_3 plate_1)

Selected rollout

LIBERO-90 · Task 012 · Episode 1

Success
Agent view observation.images.image
Eye in hand observation.images.image2
Videos are not loaded yet.

Rollout record

Episode metadata

Link to episode
Episode
1
Seed
1000
Init state
0
Control steps
122
Wall time
9.3s
Outcome
Success

Evidence browser

All rollouts

Select one episode to mount only its two camera streams in the player.

Episode Outcome Seed Init state Steps Video
1 Success 1000 0 122 168 KB
2 Success 1001 2 222 321 KB
3 Success 1002 4 112 127 KB
4 Success 1003 6 92 112 KB
5 Failure 1004 8 400 399 KB
6 Success 1005 9 83 99 KB
7 Success 1006 11 93 120 KB
8 Failure 1007 13 400 482 KB
9 Failure 1008 14 400 354 KB
10 Success 1009 15 89 115 KB
11 Success 1010 17 91 102 KB
12 Success 1011 19 127 171 KB
13 Failure 1012 21 400 313 KB
14 Success 1013 22 91 107 KB
15 Success 1014 24 265 338 KB
16 Failure 1015 26 400 339 KB
17 Success 1016 27 217 271 KB
18 Success 1017 29 254 369 KB
19 Failure 1018 31 400 406 KB
20 Failure 1019 32 400 497 KB
21 Success 1020 33 142 176 KB
22 Success 1021 35 91 118 KB
23 Success 1022 37 97 108 KB
24 Success 1023 39 151 174 KB
25 Failure 1024 41 400 323 KB
26 Failure 1025 42 400 379 KB
27 Failure 1026 43 400 342 KB
28 Failure 1027 44 400 426 KB
29 Success 1028 45 96 110 KB
30 Failure 1029 47 400 371 KB
31 Failure 1030 48 400 567 KB
32 Success 1031 49 84 110 KB
33 Failure 1032 1 400 366 KB
34 Success 1033 2 199 271 KB
35 Success 1034 4 95 115 KB
36 Success 1035 6 90 109 KB
37 Failure 1036 8 400 407 KB
38 Success 1037 9 83 100 KB
39 Success 1038 11 223 291 KB
40 Success 1039 13 91 113 KB
41 Success 1040 15 86 108 KB
42 Success 1041 17 90 97 KB
43 Success 1042 19 175 222 KB
44 Failure 1043 21 400 426 KB
45 Success 1044 22 95 105 KB
46 Success 1045 24 283 351 KB
47 Failure 1046 26 400 419 KB
48 Success 1047 27 210 281 KB
49 Success 1048 29 114 163 KB
50 Failure 1049 31 400 361 KB

Integrity

Task attestation

Status
complete
Completed
Aug 16, 2026, 6:54 AM UTC
Manifest SHA-256
aa43f3e865…85c05009
Aggregate SHA-256
1faf335e37…58ce37d1
Episode index SHA-256
c6be4b145b…d1d12b70
Validation SHA-256
8125b1afa5…7c69e9b1
Artifact inventory SHA-256
298a0b272c…1e12d2f3