Spoon In Mug¶
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✓ 20m | 09-27 15:45 | 20m | 51 | 2.7M / 19k | $0.046 | deterministic 1, grader_error 0, instance_exact 1, instance_ok 1, n_actions 201, video_rendered 1 |
| L✗ 1h 30m | 09-29 09:50 | 1h 30m | 306 | 39.2M / 171k | $0.543 | deterministic 1, grader_error 0, live_success 0, n_actions 1711, replay_success 0, video_rendered 1 |
Task instruction (upstream)
Put the metal spoon that's in the wooden bowl in the mug
No oracle video published upstream — this is the scene it starts from.
What RoboLab states about this task
| Success predicate | object_in_container |
| Predicate arguments | object: spoon_big · container: ceramic_mug · require_gripper_detached: True |
| Scored subtasks | 1 |
| Subtask predicates | pick_and_place |
| Objects | ceramic_mug glasses keyboard lizard_figurine marker remote_control smartphone wooden_bowl spoon_big computer_mouse yogurt_cup granola_bars grey_bin |
| Upstream attributes | affordance spatial |
| Upstream difficulty | moderate |
| Episode budget | 60 s |
| Other wordings | vague — Put the spoon in the mug specific — Take the metal spoon in the wooden bowl and place it inside the ceramic mug with the handle sticking out |
| Environment class | SpoonInMugTask |
| Upstream name | SpoonInMugTask |
| Defined in | robolab/tasks/benchmark/spoon_in_mug_task.py |
From https://github.com/NVlabs/RoboLab @ v0.3.1.
Tags¶
Task DomainManipulation
Skill primitives in the demo: affordancespatial
Why this task is interesting¶
Not yet written.
Capability notes¶
Not yet written.
Oracle demo review¶
Not yet reviewed.