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 006 Validated task

put the white mug on the plate and put the chocolate pudding to the right of the plate

Completed task success rate
90.0%
45 of 50
Control horizon
520
Maximum steps
Camera videos
100
Two per rollout
Evaluation time
8m 41s
Task process wall time

Canonical catalog

Task definition

Model instruction
put the white mug on the plate and put the chocolate pudding to the right of the plate
BDDL language
put the white mug on the plate and put the chocolate pudding to the right of the plate
Problem
libero_living_room_tabletop_manipulation
Objects
chocolate_pudding_1, plate_1, porcelain_mug_1, red_coffee_mug_1
Fixtures
living_room_table
Initial-state predicates
4
Fixed init states
50
Exact initial-state predicates 
(on plate_1 living_room_table_plate_init_region)
(on red_coffee_mug_1 living_room_table_red_coffee_mug_init_region)
(on chocolate_pudding_1 living_room_table_chocolate_pudding_init_region)
(on porcelain_mug_1 living_room_table_porcelain_mug_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. porcelain mug 1 is on plate 1

    (on porcelain_mug_1 plate_1)

  2. chocolate pudding 1 is on living room table plate right region

    (on chocolate_pudding_1 living_room_table_plate_right_region)
Exact parsed BDDL goal state 
(on porcelain_mug_1 plate_1)
(on chocolate_pudding_1 living_room_table_plate_right_region)

Selected rollout

LIBERO-10 · Task 006 · 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
216
Wall time
11s
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 0 216 145 KB
2 Success 1001 2 210 142 KB
3 Success 1002 4 204 137 KB
4 Success 1003 6 203 137 KB
5 Success 1004 8 213 141 KB
6 Success 1005 10 203 138 KB
7 Success 1006 12 211 142 KB
8 Failure 1007 14 520 331 KB
9 Success 1008 15 205 139 KB
10 Success 1009 17 207 138 KB
11 Success 1010 19 226 146 KB
12 Success 1011 21 201 142 KB
13 Success 1012 23 211 141 KB
14 Success 1013 25 204 140 KB
15 Success 1014 27 231 152 KB
16 Success 1015 29 198 140 KB
17 Success 1016 31 195 133 KB
18 Success 1017 33 195 139 KB
19 Success 1018 35 414 218 KB
20 Success 1019 37 210 145 KB
21 Success 1020 39 216 147 KB
22 Success 1021 41 204 137 KB
23 Success 1022 43 207 140 KB
24 Success 1023 45 247 161 KB
25 Success 1024 47 200 136 KB
26 Success 1025 49 202 137 KB
27 Success 1026 1 213 140 KB
28 Success 1027 3 206 139 KB
29 Success 1028 5 211 140 KB
30 Success 1029 7 209 144 KB
31 Failure 1030 9 520 266 KB
32 Success 1031 10 204 139 KB
33 Success 1032 12 208 141 KB
34 Failure 1033 14 520 338 KB
35 Success 1034 15 215 145 KB
36 Success 1035 17 204 137 KB
37 Success 1036 19 215 141 KB
38 Success 1037 21 201 141 KB
39 Success 1038 23 207 139 KB
40 Success 1039 25 235 150 KB
41 Success 1040 27 203 139 KB
42 Failure 1041 29 520 257 KB
43 Success 1042 30 217 141 KB
44 Success 1043 32 196 137 KB
45 Success 1044 34 205 139 KB
46 Success 1045 36 203 138 KB
47 Failure 1046 38 520 310 KB
48 Success 1047 39 204 139 KB
49 Success 1048 41 204 139 KB
50 Success 1049 43 255 168 KB

Integrity

Task attestation

Status
complete
Completed
Aug 16, 2026, 12:38 AM UTC
Manifest SHA-256
cd9a4375eb…d25f3f0f
Aggregate SHA-256
5d3f2fb088…96724951
Episode index SHA-256
794a2fba13…4d9cc407
Validation SHA-256
facbcb3356…ff89a03c
Artifact inventory SHA-256
25c4d2941c…2b3a4d06