Stack3 Rubiks Cube¶
Agent runs
Each mode of this task runs once per run. Est. cost is tokens at list price (data/prices.yml), never a bill; see the RoboLab runs for every task of a run.
Codex + GPT-6 Luna, reasoning xhigh (ChatGPT login)default run
| Trial | Started | Agent time | Model requests | Tokens in / out | Est. cost | Grade |
|---|---|---|---|---|---|---|
| U✓ 19m | 09-26 13:38 | 19m | 44 | 2.2M / 23k | $0.043 | deterministic 1, grader_error 0, instance_exact 1, instance_ok 1, n_actions 307, video_rendered 1 |
| L✓ 17m | 09-29 09:14 | 17m | 51 | 2.4M / 28k | $0.048 | deterministic 1, grader_error 0, live_success 1, n_actions 200, replay_success 1, video_rendered 1 |
Task instruction (upstream)
Stack the rubiks cubes in a tower
No oracle video published upstream — this is the scene it starts from.
What RoboLab states about this task
| Success predicate | stacked |
| Predicate arguments | objects: ['rubiks_cube', 'rubiks_cube_1', 'rubiks_cube_2'] · order: None |
| Scored subtasks | 2 |
| Subtask predicates | Subtask stacked |
| Objects | rubiks_cube rubiks_cube_1 rubiks_cube_2 |
| Upstream attributes | stacking |
| Upstream difficulty | simple |
| Episode budget | 60 s |
| Other wordings | vague — Stack everything in a tower specific — Pick up the rubiks cubes and stack them vertically on top of each other on the table |
| Environment class | Stack3RubiksCubeTask |
| Upstream name | Stack3RubiksCubeTask |
| Defined in | robolab/tasks/benchmark/rubiks_cube_stacking_task.py |
From https://github.com/NVlabs/RoboLab @ v0.3.1.
Tags¶
Task DomainManipulation
Skill primitives in the demo: stacking
Why this task is interesting¶
Not yet written.
Capability notes¶
Not yet written.
Oracle demo review¶
Not yet reviewed.