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 005 Validated task

push the plate to the front of the stove

Completed task success rate
96.0%
48 of 50
Control horizon
300
Maximum steps
Camera videos
100
Two per rollout
Evaluation time
6m 47s
Task process wall time

Canonical catalog

Task definition

Model instruction
push the plate to the front of the stove
BDDL language
push the plate to the front of the stove
Problem
libero_tabletop_manipulation
Objects
akita_black_bowl_1, cream_cheese_1, plate_1, wine_bottle_1
Fixtures
flat_stove_1, main_table, wine_rack_1, wooden_cabinet_1
Initial-state predicates
7
Fixed init states
50
Exact initial-state predicates 
(on wine_bottle_1 main_table_wine_bottle_region)
(on akita_black_bowl_1 main_table_akita_black_bowl_region)
(on plate_1 main_table_plate_region)
(on cream_cheese_1 main_table_cream_cheese_region)
(on wooden_cabinet_1 main_table_cabinet_region)
(on flat_stove_1 main_table_stove_region)
(on wine_rack_1 main_table_wine_rack_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. plate 1 is on main table stove front region

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

Selected rollout

LIBERO-Goal · Task 005 · 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
153
Wall time
9.1s
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 0 153 92 KB
2 Success 1001 2 142 93 KB
3 Success 1002 4 144 91 KB
4 Failure 1003 6 300 211 KB
5 Success 1004 7 148 95 KB
6 Success 1005 9 231 149 KB
7 Success 1006 11 238 141 KB
8 Success 1007 13 151 101 KB
9 Success 1008 15 134 90 KB
10 Success 1009 17 128 84 KB
11 Success 1010 19 149 94 KB
12 Success 1011 21 137 91 KB
13 Success 1012 23 151 99 KB
14 Success 1013 25 116 84 KB
15 Success 1014 27 122 86 KB
16 Success 1015 29 146 92 KB
17 Success 1016 31 153 92 KB
18 Success 1017 33 143 93 KB
19 Success 1018 35 138 91 KB
20 Success 1019 37 144 93 KB
21 Success 1020 39 152 90 KB
22 Success 1021 41 131 85 KB
23 Success 1022 43 163 96 KB
24 Success 1023 45 128 86 KB
25 Success 1024 47 284 202 KB
26 Success 1025 49 136 87 KB
27 Success 1026 1 146 86 KB
28 Success 1027 3 212 134 KB
29 Success 1028 5 158 96 KB
30 Success 1029 7 144 89 KB
31 Success 1030 9 155 95 KB
32 Success 1031 11 142 89 KB
33 Success 1032 13 149 95 KB
34 Success 1033 15 133 87 KB
35 Success 1034 17 150 95 KB
36 Success 1035 19 129 86 KB
37 Failure 1036 21 300 230 KB
38 Success 1037 22 149 93 KB
39 Success 1038 24 142 91 KB
40 Success 1039 26 130 88 KB
41 Success 1040 28 133 87 KB
42 Success 1041 30 151 96 KB
43 Success 1042 32 153 97 KB
44 Success 1043 34 148 94 KB
45 Success 1044 36 135 88 KB
46 Success 1045 38 127 88 KB
47 Success 1046 40 151 98 KB
48 Success 1047 42 139 90 KB
49 Success 1048 44 149 94 KB
50 Success 1049 46 116 75 KB

Integrity

Task attestation

Status
complete
Completed
Aug 16, 2026, 1:50 AM UTC
Manifest SHA-256
b363a3f241…b1013f22
Aggregate SHA-256
733052ec1f…dd3c3595
Episode index SHA-256
01fe4e28a0…5d2825d0
Validation SHA-256
e7fc77e777…5e7377dd
Artifact inventory SHA-256
cc70ab62fd…bff7e445