Model summary · Edition 1.1

MiniMax M3High

These results come from one run: 10 fixed subjects, 3 rounds each, a 40-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 30 rounds → All models

Every question, answer, guess, and source.

Rank: Rank 20of 20 models
Question score: 29.17 Lower is better. 95% confidence interval 24.71 to 33.62.
Rounds solved: 15 of 30 8 format errors
Guesser cost per round: $0.0100 Model time 1m 12s.

See how MiniMax M3 (high) played. All 30 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: 2nd of 20 models, average 7.67 questions. round 1, round 2, round 3
Bike pump: 14th of 20 models, average 38.33 questions. round 1, round 2, round 3

Lowest ranks vs other models

Spider web: 20th of 20 models, average 41.00 questions. round 1, round 2, round 3
Moon: 20th of 20 models, average 32.00 questions. round 1, round 2, round 3

Single rounds

Fewest questions in a solved round: 6, on Albert Einstein. round 1
Most questions in a solved round: 34, on Moon. round 1
Largest spread between rounds: Achilles, from 6 to 41 questions. round 1, round 2, round 3
76 wrong guesses in 30 rounds, up to 14 in one round on Garfield. round 2

Failed rounds and format errors

Not solved: Albert Schweitzer. Too many format errors after 40 questions; scored as 41. round 1
Not solved: Albert Schweitzer. Too many format errors after 40 questions; scored as 41. round 3
Not solved: Garfield. Question limit reached after 40 questions; scored as 41. round 2
Not solved: Achilles. Too many format errors after 40 questions; scored as 41. round 3
Not solved: Genghis Khan. Question limit reached after 40 questions; scored as 41. round 2
Not solved: Bike pump. Too many format errors after 40 questions; scored as 41. round 1
Not solved: Bike pump. Too many format errors after 40 questions; scored as 41. round 3
Not solved: Spider web. Question limit reached after 40 questions; scored as 41. round 1
Not solved: Spider web. Too many format errors after 40 questions; scored as 41. round 2
Not solved: Spider web. Question limit reached after 40 questions; scored as 41. round 3
Not solved: Eyebrow. Too many format errors after 40 questions; scored as 41. round 1
Not solved: Moon. Too many format errors after 40 questions; scored as 41. round 2
Not solved: Door handle. Too many format errors after 40 questions; scored as 41. round 1
Not solved: Door handle. Too many format errors after 40 questions; scored as 41. round 2
Not solved: Door handle. Question limit reached after 40 questions; scored as 41. round 3
8 format errors in 8 rounds. Each one before the question limit used a counted turn. Albert Einstein round 1, Albert Schweitzer round 3, Garfield round 3, Bike pump round 1, Moon round 1, Moon round 2, Door handle round 1, Door handle round 2

Subjects

Results by subject.

Failed rounds count as 41 questions. Rank compares this subject's average with the other models in edition 1.1. Select a subject or a round to see the game.

Subject	Average questions	Subject rank	Solved	Rounds
Albert Einstein	7.67	2 of 20	3 of 3	6, 8, 9
Albert Schweitzer	33.67	19 of 20	1 of 3	not solved, 19, not solved
Garfield	24.00	20 of 20	2 of 3	17, not solved, 14
Achilles	18.33	19 of 20	2 of 3	6, 8, not solved
Genghis Khan	24.00	19 of 20	2 of 3	12, not solved, 19
Bike pump	38.33	14 of 20	1 of 3	not solved, 33, not solved
Spider web	41.00	20 of 20	0 of 3	not solved, not solved, not solved
Eyebrow	31.67	19 of 20	2 of 3	not solved, 27, 27
Moon	32.00	20 of 20	2 of 3	34, not solved, 21
Door handle	41.00	19 of 20	0 of 3	not solved, not solved, not solved

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 1.1: BX-20260908-B-0003-experimental-M0024-001, completed 9 Sep 2026.