CookingBench

Results / v2.1 / OpenAI

Archived v2.1Unresolved leading group

GPT-5.6 Terra Pro

This profile reports GPT-5.6 Terra Pro’s original v2.1 result. It does not claim a timeless culinary rank. The run’s scoring limitations and later exploratory analyses are separate records.

Original overall

95.4

95% item-bootstrap interval

91.7–98.3

Frontier subset

91.3

Original score profile

Uncertainty and pairwise evidence

GPT-5.6 Terra Pro belonged to a five-model group from which the run’s paired item-bootstrap rule did not identify one unique leader.

Scoring limitations

Saturated numeric items, high-influence keyword items and semantic constraint-check failures affected the instrument. The profile is retained as historical evidence, not validated culinary ability.

Run 2026-07-v2.1 · openai/gpt-5.6-terra-pro · original artifact unchanged