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 001 Validated task

put both the cream cheese box and the butter in the basket

Completed task success rate
100.0%
50 of 50
Control horizon
520
Maximum steps
Camera videos
100
Two per rollout
Evaluation time
9m 22s
Task process wall time

Canonical catalog

Task definition

Model instruction
put both the cream cheese box and the butter in the basket
BDDL language
put both the cream cheese box and the butter in the basket
Problem
libero_living_room_tabletop_manipulation
Objects
alphabet_soup_1, basket_1, butter_1, cream_cheese_1, ketchup_1, milk_1, orange_juice_1, tomato_sauce_1
Fixtures
living_room_table
Initial-state predicates
8
Fixed init states
50
Exact initial-state predicates 
(on alphabet_soup_1 living_room_table_alphabet_soup_init_region)
(on cream_cheese_1 living_room_table_cream_cheese_init_region)
(on tomato_sauce_1 living_room_table_tomato_sauce_init_region)
(on ketchup_1 living_room_table_ketchup_init_region)
(on milk_1 living_room_table_milk_init_region)
(on orange_juice_1 living_room_table_orange_juice_init_region)
(on butter_1 living_room_table_butter_init_region)
(on basket_1 living_room_table_basket_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. cream cheese 1 is in basket 1 contain region

    (in cream_cheese_1 basket_1_contain_region)

  2. butter 1 is in basket 1 contain region

    (in butter_1 basket_1_contain_region)
Exact parsed BDDL goal state 
(in cream_cheese_1 basket_1_contain_region)
(in butter_1 basket_1_contain_region)

Selected rollout

LIBERO-10 · Task 001 · 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
233
Wall time
12s
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 0 233 188 KB
2 Success 1001 2 219 182 KB
3 Success 1002 4 230 195 KB
4 Success 1003 6 247 191 KB
5 Success 1004 8 248 187 KB
6 Success 1005 10 238 187 KB
7 Success 1006 12 279 227 KB
8 Success 1007 14 237 190 KB
9 Success 1008 16 220 186 KB
10 Success 1009 18 231 190 KB
11 Success 1010 20 240 189 KB
12 Success 1011 22 232 191 KB
13 Success 1012 24 227 198 KB
14 Success 1013 26 232 179 KB
15 Success 1014 28 254 219 KB
16 Success 1015 30 226 187 KB
17 Success 1016 32 235 183 KB
18 Success 1017 34 258 212 KB
19 Success 1018 36 242 202 KB
20 Success 1019 38 257 215 KB
21 Success 1020 40 234 187 KB
22 Success 1021 42 263 211 KB
23 Success 1022 44 238 209 KB
24 Success 1023 46 234 195 KB
25 Success 1024 48 248 194 KB
26 Success 1025 0 236 198 KB
27 Success 1026 2 229 187 KB
28 Success 1027 4 237 199 KB
29 Success 1028 6 247 189 KB
30 Success 1029 8 232 178 KB
31 Success 1030 10 227 184 KB
32 Success 1031 12 231 181 KB
33 Success 1032 14 237 192 KB
34 Success 1033 16 239 203 KB
35 Success 1034 18 215 183 KB
36 Success 1035 20 238 189 KB
37 Success 1036 22 239 192 KB
38 Success 1037 24 226 194 KB
39 Success 1038 26 225 175 KB
40 Success 1039 28 276 222 KB
41 Success 1040 30 219 179 KB
42 Success 1041 32 239 192 KB
43 Success 1042 34 249 199 KB
44 Success 1043 36 238 193 KB
45 Success 1044 38 259 211 KB
46 Success 1045 40 229 188 KB
47 Success 1046 42 261 211 KB
48 Success 1047 44 234 199 KB
49 Success 1048 46 242 186 KB
50 Success 1049 48 244 198 KB

Integrity

Task attestation

Status
complete
Completed
Aug 15, 2026, 11:54 PM UTC
Manifest SHA-256
f7e382d9b6…d4fa9184
Aggregate SHA-256
38376beeb6…76c60a21
Episode index SHA-256
e8e1f200a5…6d2bc644
Validation SHA-256
6ce8555b61…1b250ca9
Artifact inventory SHA-256
7348a289bc…a37ff8cd