Stand Pages¶
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 RoboWits 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 30m 7% | 09-26 01:47 | 1h 30m | 168 | 17.3M / 174k | $0.321 | deterministic 1, grader_error 0, instance_exact 1, instance_ok 1, n_actions 4970, video_rendered 1 |
| L✗ 1h 30m | 09-29 11:26 | 1h 30m | 152 | 14.1M / 91k | $0.208 | deterministic 1, grader_error 0, live_success 0, n_actions 1921, replay_success 0, video_rendered 1 |
Task instruction (upstream)
Assemble the two pages using the stabilizing bar to create a standing hinge structure.
Playback speed
What RoboWits states about this task
| Success criteria | 1. Both pages are standing upright (rotation about x or y close to 90 degrees) 2. Both pages are on the table surface 3. Both pages are within table bounds |
| Objects | nope page A with eyelets page B with eyelets stabilizing bar |
| Physics | rigid |
| Episode budget | 200 control steps |
| Default control mode | EE_ABS |
| Evaluation scenes | 50 |
| Published mutations | 1 |
| Registry id | robowits/07-stand-pages-v0 |
| Environment class | StandPagesEnv |
| Defined in | gs_gym/envs/robowits/07_stand_pages.py |
From https://github.com/UMass-Embodied-AGI/RoboWits @ 9cc30ae.
Tags¶
Task DomainManipulation
Why this task is interesting¶
Not yet written.
Capability notes¶
Not yet written.
Oracle demo review¶
Not yet reviewed.