Acropolis

Muse Wonder 1.0

MetaGrade A. Evidence grade A of A to D: strongest.ProprietaryReleased 12 Oct 2026Consistent across benchmarks
Compare with peers

Domain profile

Median within-board percentile by capability domain across boards with sufficient peer overlap.

  • Agentic (13 boards): 73 (scale 0 to 100)Agentic (13 boards)73
  • Coding (8 boards): 70 (scale 0 to 100)Coding (8 boards)70
  • Finance (5 boards): 76 (scale 0 to 100)Finance (5 boards)76
  • General text (2 boards): 99 (scale 0 to 100)General text (2 boards)99
  • Legal (3 boards): 88 (scale 0 to 100)Legal (3 boards)88

Benchmark appearances

Best verified run per board, with confidence intervals and Pareto frontier status.

BenchmarkRankScoreCostStatusObserved
LMArena Text (style control)

LMArena · Polish · max

#1 / 243Tied #1
1,546[1,505 to 1,588]UnknownStandard11 Nov 2026
LMArena Text (style control)

LMArena · Industry medicine and healthcare · max

#2 / 391Tied #1
1,523[1,502 to 1,543]UnknownStandard11 Nov 2026
LMArena Text (style control)

LMArena · Japanese · max

#2 / 288Tied #1
1,531[1,489 to 1,573]UnknownStandard11 Nov 2026
LMArena Text

LMArena · Multi turn · max

#3 / 419
1,514[1,500 to 1,529]UnknownStandard11 Nov 2026
LMArena Text

LMArena · Polish · max

#3 / 243Tied #1
1,537[1,496 to 1,577]UnknownStandard11 Nov 2026
LMArena Text (style control)

LMArena · English · max

#3 / 421Tied #1
1,519[1,510 to 1,528]UnknownStandard11 Nov 2026
LMArena Text (style control)

LMArena · Exclude ties · max

#3 / 421
1,533[1,525 to 1,542]UnknownStandard11 Nov 2026
LMArena Text (style control)

LMArena · French · max

#3 / 301Tied #1
1,526[1,498 to 1,554]UnknownStandard11 Nov 2026
LMArena Text (style control)

LMArena · Hard prompts · max

#3 / 421
1,531[1,523 to 1,539]UnknownStandard11 Nov 2026
LMArena Text (style control)

LMArena · Hard prompts English · max

#3 / 419Tied #1
1,536[1,525 to 1,548]UnknownStandard11 Nov 2026
LMArena Text (style control)

LMArena · Industry legal and government · max

#3 / 393Tied #1
1,524[1,505 to 1,544]UnknownStandard11 Nov 2026
LMArena Text (style control)

LMArena · Multi turn · max

#3 / 419
1,525[1,511 to 1,540]UnknownStandard11 Nov 2026
LMArena Text (style control)

LMArena · Non English · max

#3 / 421
1,498[1,490 to 1,505]UnknownStandard11 Nov 2026
LMArena Text (style control)

LMArena · Overall · max

#3 / 421
1,511[1,505 to 1,517]UnknownStandard11 Nov 2026
LMArena Text (style control)

LMArena · Russian · max

#3 / 385Tied #1
1,523[1,509 to 1,537]UnknownStandard11 Nov 2026
LMArena Text (style control)

LMArena · Industry business and management and financial operations · max

#4 / 414Tied #1
1,506[1,493 to 1,519]UnknownStandard11 Nov 2026
LMArena Text (style control)

LMArena · Industry software and IT services · max

#4 / 421Tied #1
1,540[1,530 to 1,549]UnknownStandard11 Nov 2026
Legal Agent Benchmark (HLAB-Vals)

Vals AI benchmarks · max

#4 / 38Tied #1
23.8%[16.6% to 30.9%]$7.08Standard15 Nov 2026
LMArena Text

LMArena · English · max

#5 / 421
1,509[1,500 to 1,519]UnknownStandard11 Nov 2026
LMArena Text

LMArena · Japanese · max

#5 / 288Tied #1
1,519[1,476 to 1,562]UnknownStandard11 Nov 2026
LMArena Text (style control)

LMArena · Industry life and physical and social science · max

#5 / 419Tied #1
1,525[1,511 to 1,539]UnknownStandard11 Nov 2026
LMArena Text (style control)

LMArena · Korean · max

#5 / 289Tied #1
1,483[1,438 to 1,528]UnknownStandard11 Nov 2026
Legal Research

Vals AI benchmarks · max

#5 / 38Tied #1
50.5%[43.7% to 57.3%]$3.98Standard15 Nov 2026
LMArena Text

LMArena · Exclude ties · max

#6 / 421
1,520[1,511 to 1,528]UnknownStandard11 Nov 2026
LMArena Text

LMArena · French · max

#6 / 301
1,512[1,484 to 1,539]UnknownStandard11 Nov 2026
LMArena Text

LMArena · Industry legal and government · max

#6 / 393Tied #1
1,520[1,500 to 1,540]UnknownStandard11 Nov 2026
LMArena Text

LMArena · Overall · max

#6 / 421
1,503[1,497 to 1,509]UnknownStandard11 Nov 2026
LMArena Text

LMArena · Russian · max

#6 / 385
1,514[1,500 to 1,528]UnknownStandard11 Nov 2026
LMArena Text (style control)

LMArena · Industry entertainment and sports and media · max

#6 / 419
1,489[1,477 to 1,500]UnknownStandard11 Nov 2026
LMArena Text (style control)

LMArena · Industry writing and literature and language · max

#6 / 420
1,499[1,489 to 1,510]UnknownStandard11 Nov 2026
LMArena Text

LMArena · Industry entertainment and sports and media · max

#7 / 419
1,485[1,474 to 1,496]UnknownStandard11 Nov 2026
LMArena Text (style control)

LMArena · Creative writing · max

#7 / 419
1,491[1,479 to 1,504]UnknownStandard11 Nov 2026
LMArena Text

LMArena · Hard prompts English · max

#9 / 419
1,519[1,507 to 1,530]UnknownStandard11 Nov 2026
LMArena Text

LMArena · Industry medicine and healthcare · max

#9 / 391Tied #1
1,506[1,485 to 1,527]UnknownStandard11 Nov 2026
LMArena Text

LMArena · Industry writing and literature and language · max

#9 / 420
1,496[1,486 to 1,507]UnknownStandard11 Nov 2026
LMArena Text

LMArena · Korean · max

#9 / 289Tied #1
1,476[1,431 to 1,520]UnknownStandard11 Nov 2026
LMArena Text

LMArena · Non English · max

#9 / 421
1,491[1,484 to 1,499]UnknownStandard11 Nov 2026
Vals Index

Vals AI benchmarks · max

#9 / 38
64.9%[62.8% to 67%]$10.65Standard15 Nov 2026
Code Migration

Vals AI benchmarks · max

#10 / 38
69.8%[61.9% to 77.7%]$47.55Standard15 Nov 2026
Finance Agent

Vals AI benchmarks · max

#10 / 38Tied #1
65.5%[61.6% to 69.5%]$2.62Standard15 Nov 2026
LMArena Text

LMArena · Creative writing · max

#10 / 419
1,493[1,481 to 1,505]UnknownStandard11 Nov 2026
LMArena Text

LMArena · Hard prompts · max

#10 / 421
1,514[1,507 to 1,522]UnknownStandard11 Nov 2026
LMArena Text

LMArena · Industry business and management and financial operations · max

#10 / 414
1,490[1,477 to 1,504]UnknownStandard11 Nov 2026
LMArena Text

LMArena · Industry software and IT services · max

#10 / 421
1,520[1,511 to 1,530]UnknownStandard11 Nov 2026
LMArena Text (style control)

LMArena · Coding · max

#10 / 416
1,542[1,531 to 1,554]UnknownStandard11 Nov 2026
LMArena Text (style control)

LMArena · Instruction following · max

#10 / 421
1,498[1,489 to 1,508]UnknownStandard11 Nov 2026
EMB

Vals AI benchmarks · max

#11 / 38Tied #1
72.5%[67.8% to 77.2%]$5.60Standard15 Nov 2026
Terminal-Bench 4

Vals AI benchmarks · max

#11 / 38
49.5%[45.2% to 53.8%]$14.66Standard15 Nov 2026
Vending-Bench 2

Epoch AI Benchmarking Hub · Default

#11 / 72
$9,432.18UnknownStandardNot stated
LMArena Text

LMArena · Instruction following · max

#12 / 421
1,494[1,485 to 1,504]UnknownStandard11 Nov 2026
LMArena Text (style control)

LMArena · Longer query · max

#12 / 399
1,507[1,499 to 1,516]UnknownStandard11 Nov 2026
LMArena Text

LMArena · Coding · max

#13 / 416
1,516[1,504 to 1,527]UnknownStandard11 Nov 2026
LMArena Text

LMArena · Industry life and physical and social science · max

#13 / 419
1,511[1,497 to 1,524]UnknownStandard11 Nov 2026
Tax Agent

Vals AI benchmarks · max

#13 / 38
36.3%[29.5% to 43.1%]$1.88Standard15 Nov 2026
LMArena Text

LMArena · Chinese · max

#14 / 395
1,536[1,517 to 1,555]UnknownStandard11 Nov 2026
LMArena Text

LMArena · Longer query · max

#15 / 399
1,497[1,488 to 1,506]UnknownStandard11 Nov 2026
LMArena Text (style control)

LMArena · Spanish · max

#15 / 304Tied #1
1,483[1,455 to 1,510]UnknownStandard11 Nov 2026
LMArena WebDev

LMArena · Webdev HTML · max

#17 / 155
1,658[1,636 to 1,680]UnknownStandard15 Nov 2026
LMArena Text (style control)

LMArena · Chinese · max

#18 / 395
1,539[1,520 to 1,557]UnknownStandard11 Nov 2026
LMArena Text (style control)

LMArena · Expert · max

#18 / 371
1,525[1,508 to 1,542]UnknownStandard11 Nov 2026
LMArena Text

LMArena · Expert · max

#19 / 371
1,514[1,497 to 1,531]UnknownStandard11 Nov 2026
Vibe Code Bench

Vals AI benchmarks · max · OpenHands

#19 / 38
83.1%[78.2% to 88%]$1.80Standard15 Nov 2026
CursorBench

CursorBench · max · Cursor

#22 / 91
50.4%$5.25StandardNot stated
LMArena Text

LMArena · Spanish · max

#22 / 304
1,477[1,450 to 1,504]UnknownStandard11 Nov 2026
LMArena WebDev

LMArena · Overall · max

#22 / 156
1,647[1,638 to 1,656]UnknownStandard15 Nov 2026
LMArena WebDev

LMArena · Webdev · max

#22 / 156
1,647[1,638 to 1,656]UnknownStandard15 Nov 2026
LMArena Text (style control)

LMArena · Industry mathematical · max

#23 / 402
1,499[1,477 to 1,522]UnknownStandard11 Nov 2026
CursorBench

CursorBench · xhigh · Cursor

#25 / 91
48.2%$4.24StandardNot stated
LMArena WebDev

LMArena · Webdev react · max

#27 / 140
1,652[1,641 to 1,664]UnknownStandard15 Nov 2026
LMArena Text (style control)

LMArena · Math · max

#29 / 404
1,490[1,467 to 1,512]UnknownStandard11 Nov 2026
LMArena Agent

LMArena · Overall · max

#30 / 64
0[-0 to 0]$1.16Standard15 Nov 2026
LMArena Text

LMArena · Industry mathematical · max

#31 / 402
1,490[1,468 to 1,513]UnknownStandard11 Nov 2026
CursorBench

CursorBench · high · Cursor

#33 / 91
46.1%$3.51StandardNot stated
LMArena Text

LMArena · Math · max

#35 / 404
1,478[1,456 to 1,501]UnknownStandard11 Nov 2026
CursorBench

CursorBench · medium · Cursor

#36 / 91
45.2%$2.82StandardNot stated
CursorBench

CursorBench · low · Cursor

#46 / 91
41.7%$1.91StandardNot stated
Bug Hunt Bench

Bug Hunt Bench · max · Muse Code

#48 / 145
31.7$27.87Standard13 Oct 2026
CursorBench

CursorBench · minimal · Cursor

#64 / 91
34.2%$0.84StandardNot stated

Reasoning effort scaling

Same benchmark and harness measured across multiple reasoning effort settings.

  • CursorBench

    CursorBench · Cursor

    Monotone. Evidence grade A of A to D: strongest.
    minimal

    34.2

    $0.84

    low

    41.7

    $1.91

    medium

    45.2

    $2.82

    high

    46.1

    $3.51

    xhigh

    48.2

    $4.24

    max

    50.4

    $5.25

Closest head-to-head rivals

Direct pairwise win rates for Muse Wonder 1.0 across shared benchmark boards.