GLM-5.3 · Codex · robocasa
GLM-5.3 · Codex: the model's API rejected the camera images the harness attached, so its vision trials failed as infrastructure failures and are not scored (31 of its 43; the other 12 because the episode server did not start). Its scored trials saw no images.
| Close the drawer (RoboCasa kitchen 1) | Solved | 112 / 400 | 1 min 5 s | |
|---|---|---|---|---|
| Close the drawer in RoboCasa kitchen layout 2 (Franka Panda on a mobile base)Step budget used up | Not solved | Step budget used up | 400 / 400 | 3 min 36 s |
| Counter to sink in RoboCasa kitchen layout 1 (Franka Panda on a mobile base)Agent said done | Not solved | Agent said done | 210 / 800 | 1 min 4 s |
| Counter to sink in RoboCasa kitchen layout 2 (Franka Panda on a mobile base) · seed 0 | Solved | 174 / 800 | 49 s | |
| Counter to sink in RoboCasa kitchen layout 2 (Franka Panda on a mobile base) · seed 2 | Solved | 186 / 800 | 58 s | |
| Counter to sink in RoboCasa kitchen layout 3 (Franka Panda on a mobile base) | Solved | 260 / 800 | 1 min 17 s | |
| Counter to sink in RoboCasa kitchen layout 4 (Franka Panda on a mobile base) | Solved | 418 / 800 | 1 min 35 s |
No trials match these filters.