Oil · Pearlearring¶
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 RoboPaint 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 53m 62% | 09-28 13:08 | 1h 53m | 175 | 24.1M / 125k | $0.331 | deterministic 1, grader_error 0, instance_exact 1, instance_ok 1, missing_trajectory 0, n_actions 70812, video_rendered 1 |
| L✗ 1h 12m 1% | 09-29 18:03 | 1h 12m | 76 | 6.7M / 113k | $0.143 | deterministic 1, grader_error 0, instance_exact 1, n_actions 34265, replay_success 0, video_rendered 1 |
Task instruction (upstream)
Paint the picture shown in /app/target.png (a square crop of Johannes Vermeer's Girl with a Pearl Earring, c. 1665) onto the main sheet with brush and oil paints, in at most 450 strokes.
Playback speed
Recorded by us: our reference solution, replayed by the verifier (canvas camera, scene camera, side camera, exact canvas).
What RoboPaint states about this task
| Success criteria | 1. scored at a 1 mm scale against the picture, all after the best single shift of the sheet, at most 5 mm along each axis, inside the central area 8 mm in from the sheet's edges 2. mean CIELAB ΔE <= 5.7 3. p90 ΔE <= 12.5 4. SSIM of L* >= 0.77 5. edge correlation >= 0.61 6. paint IoU >= 0.96 7. at most 450 strokes 8. both fresh-process replays of the trajectory meet this and end in the same state |
| Family | oil |
| Picture | a square crop of Johannes Vermeer's Girl with a Pearl Earring, c. 1665 |
| Difficulty | extreme |
| Tier | privileged |
| Continuous Score | 1 - mean ΔE / 20 |
| Grader Facts | name — pearlearring alpha — truth_alpha.npz n_strokes — 300 source — the reference picture: File:1665 Girl with a Pearl Earring.jpg (public domain), cropped and downscaled license — Public domain area_mm2 — 51612.8 |
| Scoring | s.score() is served by a separate scoring container: it replays the actions the agent executed since its last reset on the graders' copy and returns the metrics, never images or target data |
| Agent Budget | 7200 s of wall clock per mode |
From our RoboPaint task definitions.
Tags¶
Task DomainManipulation
Why this task is interesting¶
Not yet written.
Capability notes¶
Not yet written.
Oracle demo review¶
Not yet reviewed.