Skip to main content
Meo

MiniMax: MiniMax-01

MiniMax: MiniMax-01 is minimax's model. On the un-leaked meo benchmark it ranks #63 of 82 with 49.4% accuracy and an Effective Value (𝕍) of 0.0002 β€” earning a correct answer for about $0.0014 in 27s each.

as of 2026-10-18

meo rank
#63
of 82
Accuracy
49.4%
overall
𝕍 index (top=100)
0.005
raw 𝕍 0.0002 Β· N=10
$ / correct
$0.0014
27s each

Per-domain accuracy

125 correct of 250 questions asked across 11 domains Β· answer rate 100% Β· macro accuracy 45.0%. Headline accuracy is micro-averaged per meo’s methodology; per-domain question counts (correct/asked) are shown so you can judge significance.

State tracking
100%1/1
Tape machine
80%12/15
Base-rate bias
76%31/41
Framework bias
67%2/3
Theory of mind
67%2/3
Regex automaton
56%9/16
Perceptual illusions
54%55/102
Watson-Glaser
50%2/4
CSP (unsatisfiable)
25%4/16
Logic / math / CS
18%6/34
Long arithmetic
7%1/15

Efficiency

Cost per correct
$0.0014
Seconds per correct
27
Tokens per correct
672
Correct answers
125 / 250
Total benchmark cost
$0.17
Avg tokens / item
336

Model details

Released
2025-01-15
Weights
Open
Context
1.0M
Max output
900K
Knowledge cutoff
2024-03-31
Price in
$0.20/M
Price out
$1.10/M
Modality
text+image->text
View on OpenRouter β†—Hugging Face β†—

Sources: openrouter Β· meo

FAQ

How accurate is MiniMax: MiniMax-01?

MiniMax: MiniMax-01 scores 49.4% overall on the meo benchmark (rank #63 of 82), an un-leaked private holdout graded objective-first.

What is MiniMax: MiniMax-01's Effective Value?

Its Effective Value (𝕍) is 0.0002 at N=10, Ο‰=1, Ξ΄=1.5 β€” a fused metric of accuracy, speed, cost, and error-cascade. 𝕍 rankings shift with task depth.

How much does MiniMax: MiniMax-01 cost per correct answer?

About $0.0014 per correct answer (672 tokens, 27 seconds each) at benchmark-time pricing via OpenRouter.

Is MiniMax: MiniMax-01 open weights?

Yes β€” MiniMax: MiniMax-01 is released with open weights.

Benchmark updates affecting this model

  • 2026-06-09majorlogic_math_cs ground truth was broken β€” culled and rebuilt with solver-verified items

    A 5-lab frontier jury revealed that 29 of 37 LLM-authored logic_math_cs items had broken ground truth: on most, five independent frontier models unanimously concluded the clue sets were contradictory while the stored answer claimed a number, and on eight, all five agreed on the SAME different number (author arithmetic errors). Models were being marked wrong for being right. The 29 were retired (full audit published in the repo) and the domain rebuilt with generator-as-oracle items: an embedded constraint solver PROVES each puzzle has exactly one solution (or provably none), verified by an independent exhaustive enumerator. A 3-round difficulty study found frontier models saturate consistent human-scale logic puzzles, so items use anonymous global constraints and carry a measured difficulty gradient.

  • 2026-06-08majorCorrected base_bias grading (LLM-equivalence judge fix)

    The LLM-equivalence grader uses a reasoning model whose verdict token budget was too small, so its YES/NO answer was emitted as empty and silently counted as a failure. This understated the base_bias domain to ~7% for every model when the true accuracy is ~85-90%. We raised the judge budget, excluded its reasoning tokens from the verdict, switched to verbosity-robust conclusion matching, and re-graded the whole board.

Full benchmark changelog β†’

Embed this ranking

Show MiniMax: MiniMax-01’s meo rank on your site β€” the badge links back here.

Related: all MiniMax models Β· Multi-step state tracking domain Β· best Effective Value Β· biggest movers Β· methodology Β· full leaderboard.