Put Mugs On Shelf¶
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✗ 1h 00m | 09-26 04:15 | 1h 00m | 163 | 17.9M / 65k | $0.235 | grader_error 0, missing_trajectory 1 |
| L✗ 1h 30m | 09-29 09:06 | 1h 30m | 144 | 17.0M / 173k | $0.314 | deterministic 1, grader_error 0, live_success 0, n_actions 1976, replay_success 0, video_rendered 1 |
Task instruction (upstream)
Put the two mugs on the shelf
No oracle video published upstream — this is the scene it starts from.
What RoboLab states about this task
| Success predicate | object_on_top |
| Predicate arguments | object: ['ceramic_mug', 'mug'] · reference_object: rack_l04 · logical: all · require_gripper_detached: True |
| Scored subtasks | 1 |
| Subtask predicates | pick_and_place |
| Objects | ceramic_mug mug rack_l04 serving_bowl utilityjug_a01 |
| Upstream attributes | affordance spatial counting |
| Upstream difficulty | complex |
| Episode budget | 180 s |
| Other wordings | vague — Put the mugs on shelf specific — Pick up the two mugs from the table and place them on the shelf in front of you |
| Environment class | PutMugsOnShelfTask |
| Upstream name | PutMugsOnShelfTask |
| Defined in | robolab/tasks/benchmark/put_mugs_on_shelf.py |
From https://github.com/NVlabs/RoboLab @ v0.3.1.
Tags¶
Task DomainManipulation
Skill primitives in the demo: affordancecountingspatial
Why this task is interesting¶
Not yet written.
Capability notes¶
Not yet written.
Oracle demo review¶
Not yet reviewed.