Model launches / Meta

Muse Wonder and Muse Wonder Full

Scores at a glance

31 benchmarks · large dot: Muse Wonder Full; small dots: closest models where reported

Artificial Analysis Intelligence Index v5.0 ↗49.73 index

Muse Wonder Full (Max) · Intelligence Index v5.0 ↗

2 more reported results

55.80 index · Muse Wonder Full (Max) · Intelligence Index v4.3.2 · Artificial Analysis ↗

56 · Muse Wonder Full (max) · AA Index · Meta report snapshot · Muse Wonder official evaluation report ↗

AA-Briefcase v1.1 ↗1729.32 Elo

Muse Wonder Full (Max) · AA-Briefcase v1.1 · overall Elo ↗

AutomationBench-AA ↗71.33%

Muse Wonder Full (Max) · AutomationBench-AA · guardrail-gated objective score ↗

1 more reported result

45.4 · Muse Wonder Full (max) · AutomationBench-AA · Muse Wonder official evaluation report ↗

GDP.pdf ↗32.60%

Muse Wonder Full (Max) · GDP.pdf · all-pass ↗

1 more reported result

33.2 · Muse Wonder Full (max) · GDP.pdf · Muse Wonder official evaluation report ↗

Terminal-Bench 4.0 ↗57.07%

Muse Wonder Full (Max) · Terminal-Bench 4.0 · AA harness ↗

1 more reported result

57.1 · Muse Wonder Full (max) · Terminal-Bench 4.0-AA · Muse Wonder official evaluation report ↗

Terminal-Bench-Science 0.1 ↗37.14%

Muse Wonder Full (Max) · Terminal-Bench-Science 0.1 · AA harness ↗

FrontierCode 2.0 ↗37.76%

Muse Wonder Full (Max) · FrontierCode 2.0 · AA implementation ↗

CritPt v2 · hard subset ↗53.00%

Muse Wonder Full (Max) · CritPt v2 · hard subset · AA implementation ↗

HLE-Diamond ↗48.50%

Muse Wonder Full (Max) · HLE-Diamond · AA implementation ↗

AA-Omniscience v1.1 ↗30.60 index

Muse Wonder Full (Max) · AA-Omniscience v1.1 · error-penalized index ↗

1 more reported result

30.15 index · Muse Wonder Full (Max) · AA-Omniscience · error-penalized score · Artificial Analysis ↗

AA-Omniscience v1.1 · accuracy ↗40.97%

Muse Wonder Full (Max) · AA-Omniscience v1.1 · accuracy ↗

1 more reported result

43.22% · Muse Wonder Full (Max) · AA-Omniscience · accuracy · Artificial Analysis ↗

AA-Omniscience v1.1 · hallucination rate ↗17.56%

Muse Wonder Full (Max) · AA-Omniscience v1.1 · hallucination rate, lower is better ↗

1 more reported result

23.01% · Muse Wonder Full (Max) · AA-Omniscience · hallucination rate, lower is better · Artificial Analysis ↗

GDPval-AA v2.1 ↗1857.09 Elo

Muse Wonder Full (Max) · GDPval-AA v2.1 · Elo ↗

Harvey LAB-AA v1.1 ↗13.06%

Muse Wonder Full (Max) · Harvey LAB-AA v1.1 · all-pass ↗

1 more reported result

13.1 · Muse Wonder Full (max) · Harvey LAB-AA · Muse Wonder official evaluation report ↗

FrontierCode 1.1 ↗44.89%

Muse Wonder Full (Max) · FrontierCode 1.1 · AA implementation ↗

AA-LCR v1.1 ↗83.67%

Muse Wonder Full (Max) · AA-LCR v1.1 ↗

SciCode ↗62.62%

Muse Wonder Full (Max) · SciCode · AA implementation ↗

Humanity’s Last Exam ↗53.61%

Muse Wonder Full (Max) · Humanity’s Last Exam · AA implementation ↗

CritPt v2 ↗73.14%

Muse Wonder Full (Max) · CritPt v2 · code tools ↗

CritPt ↗32.86%

Muse Wonder Full (Max) · CritPt · original ↗

Personal Superintelligence Index ↗59.1

Muse Wonder Full (max) · Personal Superintelligence Index · internal ↗

Vals Finance Agent v2 ↗59.1

Muse Wonder Full (max) · Vals Finance Agent v2 ↗

HealthBench ↗62.3

Muse Wonder Full (max) · HealthBench · HealthBench Professional · length-adjusted ↗

JobBench ↗75.3

Muse Wonder Full (max) · JobBench ↗

OSWorld 2.1 ↗78.4% partial / 39.4% binary

Muse Wonder Full (max) · OSWorld 2.1 · partial and binary ↗

MemoryBench ↗85.5

Muse Wonder Full (max) · MemoryBench ↗

DeepSWE v1.1 ↗79.0

Muse Wonder Full (max) · DeepSWE 1.1 ↗

SWE-Marathon ↗57.5% · xhigh

Muse Wonder Full (max) · SWE-Marathon ↗

ProgramBench ↗59.0

Muse Wonder Full (max) · ProgramBench ↗

FocusBench ↗54.2%

Muse Wonder Full (max) · FocusBench · score · Meta · Muse Wonder launch · maximum effort ↗

GDPx v1.0 ↗9.0%

Muse Wonder Full · GDPx v1.0 · Overall · intended grading ↗

53 more reported results

6.2% · Muse Wonder Full · GDPx v1.0 · Eight-domain subset · intended grading · GDPx v1.0 — SemiAnalysis ↗

15.8% · Muse Wonder Full · GDPx v1.0 · Recursive Self-Improvement · intended grading · GDPx v1.0 — SemiAnalysis ↗

0.6% · Muse Wonder Full · GDPx v1.0 · Chip Design · intended grading · GDPx v1.0 — SemiAnalysis ↗

35.2% · Muse Wonder Full · GDPx v1.0 · Cybersecurity · intended grading · GDPx v1.0 — SemiAnalysis ↗

5.1% · Muse Wonder Full · GDPx v1.0 · Biology · intended grading · GDPx v1.0 — SemiAnalysis ↗

0.0% · Muse Wonder Full · GDPx v1.0 · Finance · intended grading · GDPx v1.0 — SemiAnalysis ↗

0.0% · Muse Wonder Full · GDPx v1.0 · Legal · intended grading · GDPx v1.0 — SemiAnalysis ↗

18.3% · Muse Wonder Full · GDPx v1.0 · Quant Trading · intended grading · GDPx v1.0 — SemiAnalysis ↗

1.6% · Muse Wonder Full · GDPx v1.0 · Hardware Design · intended grading · GDPx v1.0 — SemiAnalysis ↗

10.9% · Muse Wonder Full · GDPx v1.0 · Materials Science · intended grading · GDPx v1.0 — SemiAnalysis ↗

9.2% · Muse Wonder Full · GDPx v1.0 · Alignment · intended grading · GDPx v1.0 — SemiAnalysis ↗

2.2% · Muse Wonder Full · GDPx v1.0 · Software Engineering · intended grading · GDPx v1.0 — SemiAnalysis ↗

0.7% · Muse Wonder Full · GDPx v1.0 · Aria · Chip Design · intended grading · GDPx v1.0 — SemiAnalysis ↗

0.0% · Muse Wonder Full · GDPx v1.0 · Arcverex · Chip Design · intended grading · GDPx v1.0 — SemiAnalysis ↗

1.0% · Muse Wonder Full · GDPx v1.0 · Ergodex · Chip Design · intended grading · GDPx v1.0 — SemiAnalysis ↗

35.2% · Muse Wonder Full · GDPx v1.0 · Arithmetic · Cybersecurity · intended grading · GDPx v1.0 — SemiAnalysis ↗

5.1% · Muse Wonder Full · GDPx v1.0 · LatchBio · Biology · intended grading · GDPx v1.0 — SemiAnalysis ↗

0.0% · Muse Wonder Full · GDPx v1.0 · Living Context · Finance · intended grading · GDPx v1.0 — SemiAnalysis ↗

0.0% · Muse Wonder Full · GDPx v1.0 · Chakra · Legal · intended grading · GDPx v1.0 — SemiAnalysis ↗

18.3% · Muse Wonder Full · GDPx v1.0 · Third Moment · Quant Trading · intended grading · GDPx v1.0 — SemiAnalysis ↗

15.8% · Muse Wonder Full · GDPx v1.0 · Extrapolate · Recursive Self-Improvement · intended grading · GDPx v1.0 — SemiAnalysis ↗

1.6% · Muse Wonder Full · GDPx v1.0 · Beautiful Machines · Hardware Design · intended grading · GDPx v1.0 — SemiAnalysis ↗

9.9% · Muse Wonder Full · GDPx v1.0 · ValenceForge · Materials Science · intended grading · GDPx v1.0 — SemiAnalysis ↗

11.9% · Muse Wonder Full · GDPx v1.0 · Emulated · Materials Science · intended grading · GDPx v1.0 — SemiAnalysis ↗

9.2% · Muse Wonder Full · GDPx v1.0 · Fleet · Alignment · intended grading · GDPx v1.0 — SemiAnalysis ↗

2.2% · Muse Wonder Full · GDPx v1.0 · Datacurve · Software Engineering · intended grading · GDPx v1.0 — SemiAnalysis ↗

17.2% · Muse Wonder Full · GDPx v1.0 · Overall · partial credit · GDPx v1.0 — SemiAnalysis ↗

16.8% · Muse Wonder Full · GDPx v1.0 · Eight-domain subset · partial credit · GDPx v1.0 — SemiAnalysis ↗

15.8% · Muse Wonder Full · GDPx v1.0 · Recursive Self-Improvement · partial credit · GDPx v1.0 — SemiAnalysis ↗

2.7% · Muse Wonder Full · GDPx v1.0 · Chip Design · partial credit · GDPx v1.0 — SemiAnalysis ↗

35.2% · Muse Wonder Full · GDPx v1.0 · Cybersecurity · partial credit · GDPx v1.0 — SemiAnalysis ↗

11.1% · Muse Wonder Full · GDPx v1.0 · Biology · partial credit · GDPx v1.0 — SemiAnalysis ↗

35.8% · Muse Wonder Full · GDPx v1.0 · Finance · partial credit · GDPx v1.0 — SemiAnalysis ↗

38.4% · Muse Wonder Full · GDPx v1.0 · Legal · partial credit · GDPx v1.0 — SemiAnalysis ↗

18.3% · Muse Wonder Full · GDPx v1.0 · Quant Trading · partial credit · GDPx v1.0 — SemiAnalysis ↗

1.6% · Muse Wonder Full · GDPx v1.0 · Hardware Design · partial credit · GDPx v1.0 — SemiAnalysis ↗

19.3% · Muse Wonder Full · GDPx v1.0 · Materials Science · partial credit · GDPx v1.0 — SemiAnalysis ↗

9.2% · Muse Wonder Full · GDPx v1.0 · Alignment · partial credit · GDPx v1.0 — SemiAnalysis ↗

2.2% · Muse Wonder Full · GDPx v1.0 · Software Engineering · partial credit · GDPx v1.0 — SemiAnalysis ↗

7.1% · Muse Wonder Full · GDPx v1.0 · Aria · Chip Design · partial credit · GDPx v1.0 — SemiAnalysis ↗

0.0% · Muse Wonder Full · GDPx v1.0 · Arcverex · Chip Design · partial credit · GDPx v1.0 — SemiAnalysis ↗

1.0% · Muse Wonder Full · GDPx v1.0 · Ergodex · Chip Design · partial credit · GDPx v1.0 — SemiAnalysis ↗

35.2% · Muse Wonder Full · GDPx v1.0 · Arithmetic · Cybersecurity · partial credit · GDPx v1.0 — SemiAnalysis ↗

11.1% · Muse Wonder Full · GDPx v1.0 · LatchBio · Biology · partial credit · GDPx v1.0 — SemiAnalysis ↗

35.8% · Muse Wonder Full · GDPx v1.0 · Living Context · Finance · partial credit · GDPx v1.0 — SemiAnalysis ↗

38.4% · Muse Wonder Full · GDPx v1.0 · Chakra · Legal · partial credit · GDPx v1.0 — SemiAnalysis ↗

18.3% · Muse Wonder Full · GDPx v1.0 · Third Moment · Quant Trading · partial credit · GDPx v1.0 — SemiAnalysis ↗

15.8% · Muse Wonder Full · GDPx v1.0 · Extrapolate · Recursive Self-Improvement · partial credit · GDPx v1.0 — SemiAnalysis ↗

1.6% · Muse Wonder Full · GDPx v1.0 · Beautiful Machines · Hardware Design · partial credit · GDPx v1.0 — SemiAnalysis ↗

23.3% · Muse Wonder Full · GDPx v1.0 · ValenceForge · Materials Science · partial credit · GDPx v1.0 — SemiAnalysis ↗

15.3% · Muse Wonder Full · GDPx v1.0 · Emulated · Materials Science · partial credit · GDPx v1.0 — SemiAnalysis ↗

9.2% · Muse Wonder Full · GDPx v1.0 · Fleet · Alignment · partial credit · GDPx v1.0 — SemiAnalysis ↗

2.2% · Muse Wonder Full · GDPx v1.0 · Datacurve · Software Engineering · partial credit · GDPx v1.0 — SemiAnalysis ↗

Closest models by score

Similar positions on shared benchmark leaderboards. Smaller gaps are closer.

Muse Wonder →Meta

15.0 percentile-point gap

20 of 20 shared benchmark comparisons

Kimi K3.1 Pro →Kimi

15.4 percentile-point gap

18 of 20 shared benchmark comparisons

Claude Sonnet 5.5 →Anthropic

15.9 percentile-point gap

20 of 20 shared benchmark comparisons

How similarity is measured

Mean absolute gap in percentile positions on shared benchmark settings; lower is closer. At least 60% overlap is required, with a minimum of three when available. Independent evaluations are used when there are at least three; otherwise published report comparisons are used. Repeated model observations within one setting are averaged before ranking. Writing-style traits and composite indices are excluded. This describes similarity, not an overall quality ranking.

About the model report

The published suite includes personal work, memory, coding, and an internal composite. Partial and binary computer-use scores are separately labeled. Reviewed 2026-10-25.

Read the official model report ↗

15 benchmarks in the report

Search this collection →

RSN / BENCHMARK66 reported results

Artificial Analysis Intelligence Index v5.0

Nine evaluations connect knowledge work, coding, science and general capability in one versioned score.

Weighted composite · index points↗

AGT / BENCHMARK39 reported results

AutomationBench-AA

Agents orchestrate workflows in simulated SaaS systems while respecting operational constraints.

Objectives completed↗

KNW / BENCHMARK29 reported results

Harvey LAB-AA v1.1

Legal work is assessed against task criteria and the supplied source documents, with a gate for material hallucinations.

Hallucination-gated all-pass↗

CTX / BENCHMARK103 reported results

GDP.pdf

Professional questions require reasoning over long PDF documents and meeting task-specific requirements.

All-pass rate↗

COD / BENCHMARK115 reported results

Terminal-Bench 4.0

Complex terminal work spans coding, science, infrastructure, and other technical tasks, with executable checks.

Verified task success↗

KNW / BENCHMARK30 reported results

HealthBench

Physician-authored rubrics evaluate health conversations. Original HealthBench and the Professional evaluation use distinct measures and settings.

Rubric score↗

COD / BENCHMARK97 reported results

DeepSWE v1.1

Original engineering problems test coding agents on substantial changes in existing repositories.

Verified task success↗

COD / BENCHMARK11 reported results

SWE-Marathon

Extended software projects test whether an agent can keep making useful progress through long engineering sessions.

Long-horizon task score↗

COD / BENCHMARK81 reported results

ProgramBench

Agents reconstruct programs from observed behavior rather than receiving a conventional repository repair issue.

Behavioral reconstruction score↗

AGT / BENCHMARK13 reported results

OSWorld 2.1

The OSWorld version reported in recent launches measures computer-use agents on extended application workflows.

Partial and binary task score↗

FIN / BENCHMARK96 reported results

Vals Finance Agent v2

Agentic financial research evaluates whether a system can use tools and evidence to complete analyst work.

Financial research task score↗

AGT / BENCHMARK5 reported results

JobBench

Meta’s reported professional-tool-use evaluation measures a slice of work performed through tools.

Professional tool-use score↗

CTX / BENCHMARK5 reported results

MemoryBench

Meta reports a benchmark for retaining and retrieving useful context in model-driven workflows.

Context-memory score↗

AGT / BENCHMARK5 reported results

FocusBench

Meta uses this evaluation to study whether an agent continues working toward a difficult objective.

Sustained-focus score↗

AGT / BENCHMARK5 reported results

Personal Superintelligence Index

Meta’s internal composite summarizes personal and knowledge-work capability.

Internal composite↗

COMPARISON CHECKLIST

Match benchmark version, task subset, model effort, harness, tool access, attempt budget, scoring rule, and fallback behavior. A shared benchmark name is the start of a comparison, not proof that two reported scores use the same contract.

Meta original report ↗