Decision model

GPT-6 Luna

Decisions API, beta · 1 question per request

63.6% of the 2,198 questions right (61.6% to 65.6%, 95% interval). Back to the leaderboard.

How we called it

Kind

Decision model

Access

OpenAI Decisions API (public beta)

Paid as

Provider service

Model ID

gpt-6-luna

Variant

Decisions API, beta

Confidence from

Option score

Run

143, 8 October 2026

Settings

Reasoning

off

Request format

1 question per request

Requests at once

4

Temperature

not recorded

Harness seed

0

Harness commit

not recorded

Prompt version

not recorded

Results

Comparison group: rstuv. Compare with another setup.

4 questions got no valid answer; they count as wrong.

Page images (preliminary)

Preliminary: these numbers are not final and may change. This setup is condition A of the Page images tab, which asks the 763 questions with underlined or numbered words as text (A), as a page image (B), as both (C) and through OCR models (D); the share answered right, with its 95% interval.

Confidence

Answers kept at 99% or more

368, of which 3 wrong (share right: 97.6% to 99.7%, 95% interval); 1,830 skipped: 1,826 with the model less than 99% sure and 4 without a valid answer

Calibration gap

2.4 points

AUROC, all questions

0.82 0.80 to 0.84

AUROC, reading

0.86 0.82 to 0.89

AUROC, grammar

0.77 0.74 to 0.79

Brier score

0.166

Accuracy of its surest answers

most confident fifth: 98.6% right; most confident half: 86.6% right; most confident four fifths: 72.8% right

Typical score of a right answer

83.0%

Typical score of a wrong answer

39.5%

Repeats

We did not repeat this setup.

Latency

Latency, median request

0.191 s (this run)

Latency, 90th percentile

0.250 s

Requests at once

4

Throughput

1214.8 questions per minute of request time

Request time

1.8 minutes, over the requests whose answers were kept

Latency: the time from sending one request to receiving its answer, measured from our computer, network and any retries within the request included. Request time: the time at least one request was in flight, over the requests whose answers were kept. It is not the run's elapsed time: it leaves out requests that failed and any pause before a run was resumed.

Cost

$ per 1,000 questions

$0.040

Whole run

$0.089

Basis

Estimated from published prices and recorded token usage

Price

$0.10 input, output free per 1M tokens (source, 2026-10-08)

Input tokens per question

404

Output tokens per question

0

Going first in front of an LLM

Replay of recorded answers, cutoff chosen on half the questions, averaged over the splits. Savings are estimates from API prices; the Method page explains how.