MolmoAct 2 Evaluation gallery
IN PROGRESS · PROVISIONAL Only fully completed, validated task shards are shown. Campaign coverage and aggregate rates can change.
Paper protocol Task 002 Validated task

pick up the black bowl from table center and place it on the plate

BDDL language: pick the akita black bowl from table center and place it on the plate

Completed task success rate
98.0%
49 of 50
Control horizon
280
Maximum steps
Camera videos
100
Two per rollout
Evaluation time
6m 17s
Task process wall time

Canonical catalog

Task definition

Model instruction
pick up the black bowl from table center and place it on the plate
BDDL language
pick the akita black bowl from table center and place it on the plate
Problem
libero_tabletop_manipulation
Objects
akita_black_bowl_1, akita_black_bowl_2, cookies_1, glazed_rim_porcelain_ramekin_1, plate_1
Fixtures
flat_stove_1, main_table, wooden_cabinet_1
Initial-state predicates
7
Fixed init states
50
Exact initial-state predicates 
(on akita_black_bowl_1 main_table_table_center)
(on akita_black_bowl_2 main_table_next_to_plate_region)
(on plate_1 main_table_plate_region)
(on cookies_1 main_table_box_region)
(on glazed_rim_porcelain_ramekin_1 main_table_ramekin_region)
(on wooden_cabinet_1 main_table_cabinet_region)
(on flat_stove_1 main_table_stove_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 1 is on plate 1

    (on akita_black_bowl_1 plate_1)
Exact parsed BDDL goal state 
(on akita_black_bowl_1 plate_1)

Selected rollout

LIBERO-Spatial · Task 002 · Episode 1

Failure
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
280
Wall time
13s
Outcome
Failure

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 Failure 1000 0 280 190 KB
2 Success 1001 1 95 86 KB
3 Success 1002 3 87 84 KB
4 Success 1003 5 95 86 KB
5 Success 1004 7 99 93 KB
6 Success 1005 9 90 91 KB
7 Success 1006 11 100 92 KB
8 Success 1007 13 89 88 KB
9 Success 1008 15 95 87 KB
10 Success 1009 17 98 94 KB
11 Success 1010 19 94 85 KB
12 Success 1011 21 91 85 KB
13 Success 1012 23 105 90 KB
14 Success 1013 25 86 84 KB
15 Success 1014 27 86 88 KB
16 Success 1015 29 96 92 KB
17 Success 1016 31 92 87 KB
18 Success 1017 33 97 85 KB
19 Success 1018 35 88 87 KB
20 Success 1019 37 95 95 KB
21 Success 1020 39 96 87 KB
22 Success 1021 41 92 89 KB
23 Success 1022 43 95 89 KB
24 Success 1023 45 89 86 KB
25 Success 1024 47 132 136 KB
26 Success 1025 49 102 86 KB
27 Success 1026 1 95 85 KB
28 Success 1027 3 89 86 KB
29 Success 1028 5 90 82 KB
30 Success 1029 7 92 83 KB
31 Success 1030 9 90 82 KB
32 Success 1031 11 102 94 KB
33 Success 1032 13 93 86 KB
34 Success 1033 15 97 89 KB
35 Success 1034 17 96 98 KB
36 Success 1035 19 97 87 KB
37 Success 1036 21 89 81 KB
38 Success 1037 23 103 89 KB
39 Success 1038 25 87 82 KB
40 Success 1039 27 83 81 KB
41 Success 1040 29 100 94 KB
42 Success 1041 31 93 88 KB
43 Success 1042 33 98 87 KB
44 Success 1043 35 90 91 KB
45 Success 1044 37 92 86 KB
46 Success 1045 39 93 84 KB
47 Success 1046 41 88 89 KB
48 Success 1047 43 91 84 KB
49 Success 1048 45 93 88 KB
50 Success 1049 47 81 82 KB

Integrity

Task attestation

Status
complete
Completed
Aug 16, 2026, 3:39 AM UTC
Manifest SHA-256
cee83303e1…084a0984
Aggregate SHA-256
e70eb61371…b52c77ac
Episode index SHA-256
47fceca2d1…e27d4324
Validation SHA-256
109ed81245…4cdc7dff
Artifact inventory SHA-256
4dc35c7249…4a595c0d