BenchLeader

Muse Wonder Full

Muse Wonder Full is a Meta proprietary reasoning model, released 12 Oct 2026. Its best configuration (max reasoning effort) ranks #3 of 775 on the BenchLeader Index at 70.7 ±8.8, in the top ten. The ± is the point: 89 other configurations score within that range, so they and this one cannot be told apart on quality alone — price and speed are what separate them. It scores highest in reasoning (95) and lowest in long context (67). Last measured 19 Oct 2026.

Blended price
–
Output speed
–
First answer
–
Context
1M
Full answer
–
Cost per run
–
Released
12 Oct 2026
How it scores by categoryDashed line = average model (50). One step of 15 = one standard deviation.
  1. Overall index71
  2. Reasoning95
  3. Coding72
  4. Knowledge71
  5. Long context67
  6. Composite92

How its index has moved

16 Oct 2026 to 20 Oct 2026
607080

The index is recomputed from scratch every day, so a line moves when a new benchmark result lands, when a publisher revises a score, or when the models it is normalised against change. Early movement usually means the score is still settling.

Benchmark results

See all LLM benchmarks. One column per reasoning effort. Rank is among every configuration of every model on that benchmark. Hover a score for the run it came from.

Reasoning

BenchmarkmaxSourceTrend
Humanity's Last Exam (AA)not in index53.6%#16Artificial Analysis
CritPt32.9%#1Artificial Analysis
MysteryMechanismnot in index35.1%#10Vals AI

Coding

BenchmarkmaxSourceTrend
Arena WebDev1726#6Arena (LMArena)
SciCode (AA)not in index62.6%#4Artificial Analysis
IOI79.8%#11Vals AI
Code Migrationnot in index61.0%#7Vals AI
ProgramBenchnot in index0.5%#18Vals AI
Vibe Code Bench 1-100not in index34.0%#1Vals AI
Vibe Code Bench v1.1not in index86.0%#13Vals AI
Terminal-Bench 4.0 (AA)not in index57.1%#6Artificial Analysis

Agents & tools

BenchmarkmaxSourceTrend
GDPval-AA v2.1not in index67.8%#2Artificial Analysis
CyberBenchnot in index73.6%#11Vals AI
Finance Agent v2not in index59.1%#6Vals AI
Harvey's Legal Agent Benchmarknot in index23.3%#3Vals AI
Legal Research Benchnot in index53.4%#5Vals AI
SREBenchnot in index11.8%#12Vals AI
Tax Agent Benchnot in index43.0%#6Vals AI
Terminal-Bench 4.0 (Vals)not in index58.1%#4Vals AI
Terminal-Bench Sciencenot in index37.1%#7Vals AI
AutomationBenchnot in index71.3%#4Artificial Analysis
GDP.pdf (AA)not in index32.6%#1Artificial Analysis
Harvey LAB-AA v1.1not in index13.1%#1Harvey
AA-Briefcase v1.1not in index1729#5Artificial Analysis

Knowledge

BenchmarkmaxSourceTrend
AA-Omniscience30.1#32Artificial Analysis
LegalBench87.4%#4Vals AI
BioMysteryBenchnot in index73.3%#8Vals AI
Excel Modeling Benchmarknot in index71.9%#9Vals AI
MedCodenot in index49.9%#18Vals AI
MedScribenot in index89.3%#8Vals AI
Public Benefits Benchnot in index73.8%#3Vals AI
AA-Omniscience: accuracynot in index43.2%#99Artificial Analysis
AA-Omniscience: non-hallucinationnot in index77.0%#17Artificial Analysis

Long context

BenchmarkmaxSourceTrend
AA-LCR83.7%#15Artificial Analysis

Composite

BenchmarkmaxSourceTrend
AA Intelligence Index v4.3.255.8#4Artificial Analysis
Vals Indexnot in index66.8#4Vals AI

Where it wins

Benchmarks where this configuration ranks in the top five of every configuration measured.

More from Meta

New this month

Compare Muse Wonder Full with…

See also

Data as of 21 Oct 2026. Compare with another model.

Cite as: BenchLeader, “Muse Wonder Full: benchmarks, pricing, speed and rank”, https://www.benchleader.com/models/muse-wonder-full, data as of 21 Oct 2026.