gpt-5.6-terra Benchmark & Insights

OpenAI OpenAI API

Updated Jul 18, 2026 All models

Sample size

157 runs

in window

Accuracy

66.9%

consensus match · 160d

Confidence

92%

over 162 runs

Window end

Jul 18, 2026

most recent run

Input price

$2.50/MTok

prompt tokens

Output price

$15.00/MTok

completion tokens

Model insights

01 The black sheep of the gpt-5.6 trio: 24 points behind luna despite sharing a family, with 74 overratings vs 1 under and about 30 false "unsafe" days.
02 At $2.50/$15 it costs mid-tier money for bottom-tier judgment; its late-window behavior (four false "unsafe" calls in the last three weeks) shows no improvement.

Recent forecasts

Date

Conf.

Risk

Safe