codex-gpt-5.5 [medium]
Blind track
2.25/4
Overall
4
Families scored
12
Seeds
0
Infra errors
±0.47
Std error
Per-task family scores
Failure-level histogram
L0 (failed L1)0
L1 structural5
L2 geometric0
L3 physical constraints0
L4 dfm7
infra error0
Efficiency & cost
| Metric | Value | Source |
|---|---|---|
| Agent calls | not available | opaque / not reported |
| Wall time (s) | not available | opaque / not reported |
| Cost (USD) | not available | opaque / not reported |
| Total tokens | not available | opaque / not reported |
Perception track
2.50/4
Overall
4
Families scored
12
Seeds
0
Infra errors
±0.50
Std error
Per-task family scores
Failure-level histogram
L0 (failed L1)0
L1 structural4
L2 geometric0
L3 physical constraints0
L4 dfm8
infra error0
Efficiency & cost
| Metric | Value | Source |
|---|---|---|
| Agent calls | not available | opaque / not reported |
| Wall time (s) | not available | opaque / not reported |
| Cost (USD) | not available | opaque / not reported |
| Total tokens | not available | opaque / not reported |
Perception summary
| Observations | 13 |
| Compiled | 12 |
| Render artifacts | 36 |
| Section artifacts | 0 |
| Warnings | 4 |
| Mean iterations | 2.08 |