Model summary · Edition 1.0

GPT-5.6 LunaHigh

These results come from one run: 7 fixed subjects, 5 rounds each, a 50-question limit, and reasoning setting high. They do not predict results on other subjects or settings.

This page always shows the model’s newest ranked run.

See all 35 rounds → All models

Every question, answer, guess, and source.

Rank: Rank 16of 18 models
Question score: 17.74 Lower is better. 95% confidence interval 14.74 to 20.75.
Rounds solved: 32 of 35 1 format error
Guesser cost per round: $0.0184 Model time 1m 17s.

See how GPT-5.6 Luna (high) played. All 35 rounds, with every question, answer, guess, and source.

Observations

What stood out in this run.

Counts and ranks from this run only. A subject rank compares this model’s average on that subject with every other model in this edition. Each item links to its rounds.

Best ranks vs other models

Albert Einstein: 6th of 18 models, average 9.60 questions. round 1, round 2, round 3, round 4, round 5
Mario: 11th of 18 models, average 7.00 questions. round 1, round 2, round 3, round 4, round 5

Lowest ranks vs other models

Garfield: 16th of 18 models, average 26.80 questions. round 1, round 2, round 3, round 4, round 5
Albert Schweitzer: 14th of 18 models, average 37.40 questions. round 1, round 2, round 3, round 4, round 5

Single rounds

Fewest questions in a solved round: 5, on Mario. round 2
Most questions in a solved round: 35, on Albert Schweitzer. round 4
Largest spread between rounds: Garfield, from 16 to 51 questions. round 1, round 2, round 3, round 4, round 5
4 wrong guesses in 35 rounds, never more than one per round. round 5

Failed rounds and format errors

Not solved: Albert Schweitzer. Too many format errors after 50 questions; scored as 51. round 2
Not solved: Albert Schweitzer. Too many format errors after 50 questions; scored as 51. round 3
Not solved: Garfield. Too many format errors after 50 questions; scored as 51. round 2
1 format error in 1 round. Each one before the question limit used a counted turn. Albert Schweitzer round 2

Subjects

Results by subject.

Failed rounds count as 51 questions. Rank compares this subject's average with the other models in edition 1.0. Select a subject or a round to see the game.

Subject	Average questions	Subject rank	Solved	Rounds
Albert Einstein	9.60	6 of 18	5 of 5	7, 11, 9, 13, 8
Albert Schweitzer	37.40	14 of 18	3 of 5	20, not solved, not solved, 35, 30
Stephen King	19.00	14 of 18	5 of 5	13, 24, 13, 18, 27
Garfield	26.80	16 of 18	4 of 5	25, not solved, 21, 16, 21
Achilles	9.20	12 of 18	5 of 5	7, 8, 7, 8, 16
Genghis Khan	15.20	12 of 18	5 of 5	11, 14, 13, 21, 17
Mario	7.00	11 of 18	5 of 5	6, 5, 6, 13, 5

Runs

Runs of this model.

Each edition ranks only the newest complete run of a model. Earlier runs are not ranked and their details are not published.

Current run in edition 1.0: BX-20260728-official-M0001-010, completed 28 Jul 2026.
Edition 1.1: rank 15 of 20, question score 21.03. Different rules and subjects, so scores are not comparable across editions.