RankingEffort benchmarks

Muse Wonder Full

Compare models

19 measured results · 1 effort settings · 19 benchmark/harness combinations

Model evidence and sources →

Change model

Model

Current configurationSources

Compare with

Choose a second model and its effort to see both profiles and benchmark differences below.

Compare withModel and configuration

Capability profile

One configuration, seven capabilities

Comparison score · 0–100

AMuse Wonder Full · Max3 of 7 capabilities supported

Gaps are unknown, not zero. Hollow points are preliminary.

Agentic
A86.2Supported
Hard reasoning
A69.1Supported
Coding
A84.6Supported
Human pref
Unknown
Knowledge
A37.8Preliminary
Multimodal
Unknown
Long context
A70.6Preliminary

Each profile belongs to the named configuration. Capability scores use a shared panel for all axes, distinct from the overall ranking panel; they are not accuracy percentages or direct head-to-head wins.

Capability score evidence

Muse Wonder Full · Max: 85.0 overall comparison score. Meets the overall evidence requirements.

5 areas measured. The available task mix affects this estimate; missing areas remain unknown.

Which sources carry the weight?

Artificial Analysis · 69.8% direct comparison weight · independent of the model provider
Vals AI · 30.2% direct comparison weight · independent of the model provider

Which benchmark families carry the weight?

harvey · 9.0% direct comparison weight
Humanity’s Last Exam · 8.0% direct comparison weight
Critpt · 8.0% direct comparison weight
omniscience · 8.0% direct comparison weight
Aa Lcr · 7.6% direct comparison weight
Other families (11) · 59.5% direct comparison weight

These shares describe direct comparison weight, not fractions of the final score or confidence. Source balancing does not give every family equal weight or remove correlations between overlapping sources.

How sensitive is this score?

The audit removes one source at a time and refits against the same overall reference panel. These published results use the matching dataset audited 2026-10-20.

Both ranks use the same surviving configurations for each removal. The cohort starts with 444 configurations supported in the original audit, including separate effort settings. These are diagnostic positions, not the default leaderboard ranks. Eligibility is not rechecked after removal.

What changes when a source is removed?
Removed source	Rank before → after	Rank movement	Cohort remaining / lost support	Score after (change)
Artificial Analysis	13 → 14	Down 1	203 remaining / 241 lost	87.1 (+2.1 points)
ARC Prize	13 → 14	Down 1	444 remaining / 0 lost	87.2 (+2.2 points)
Datacurve	13 → 13	No movement	444 remaining / 0 lost	83.5 (-1.6 points)
LiveCodeBench	13 → 13	No movement	444 remaining / 0 lost	85.0 (-0.0 points)
Scale	13 → 14	Down 1	444 remaining / 0 lost	85.4 (+0.4 points)
Terminal-Bench	13 → 13	No movement	444 remaining / 0 lost	85.1 (+0.1 points)
Cognition	13 → 13	No movement	444 remaining / 0 lost	84.7 (-0.3 points)
LiveBench	13 → 13	No movement	443 remaining / 1 lost	86.6 (+1.6 points)
SWE-rebench	13 → 13	No movement	444 remaining / 0 lost	84.8 (-0.2 points)
Epoch AI	13 → 13	No movement	431 remaining / 13 lost	85.2 (+0.2 points)
Vals AI	13 → 18	Down 5	443 remaining / 1 lost	81.6 (-3.4 points)
Provider reports · Anthropic	13 → 12	Up 1	444 remaining / 0 lost	84.9 (-0.1 points)
Provider reports · Moonshot AI	13 → 13	No movement	444 remaining / 0 lost	84.8 (-0.3 points)
Provider reports · Meta	13 → 13	No movement	444 remaining / 0 lost	84.7 (-0.3 points)
Provider reports · Mistral	13 → 13	No movement	444 remaining / 0 lost	85.0 (+0.0 points)
Provider reports · OpenAI	13 → 12	Up 1	444 remaining / 0 lost	84.9 (-0.1 points)
All provider reports	13 → 11	Up 2	444 remaining / 0 lost	84.2 (-0.8 points)

Lost support means a configuration no longer has a comparison path to the complete reference panel. Its rank is unavailable. A surviving path alone does not establish leaderboard eligibility.

Overlapping sources may remain correlated even with source balancing. Leave-one-source-out movement is a sensitivity diagnostic, not a spread of source-specific ranks or a confidence measure. Scoring methodology · Dated configuration audit

Every benchmark family

Removed evidence	Recomputed score	Change
omniscience	87.3	+2.3
harvey	83.2	-1.8
deepswe	83.5	-1.5
arc-agi	86.4	+1.4
hle	85.6	+0.5
vibe-code	85.5	+0.5
briefcase	84.5	-0.5
automationbench	84.5	-0.5
aa-lcr	85.5	+0.5
vals-tax-agent	84.6	-0.4
mmmu-pro	85.4	+0.4
vals-legal-research	84.6	-0.4
vals-finance-agent	84.7	-0.3
gdpval	84.7	-0.3
gdp-pdf	84.7	-0.3
frontiercode	84.7	-0.3
arc-agi-3	84.7	-0.3
vals-excel-modeling	85.3	+0.3
vals-code-migration	85.3	+0.3
ifbench	84.8	-0.3
epoch-game-puzzles	84.8	-0.2
swe-rebench	84.8	-0.2
multi-swe-bench	84.8	-0.2
tau-bench	84.8	-0.2
mmlu-pro	85.2	+0.2
livebench-language	85.2	+0.2
swe-atlas-test-writing	84.8	-0.2
simpleqa	85.2	+0.2
critpt	85.2	+0.2
gpqa	84.9	-0.1
swe-atlas-qna	84.9	-0.1
itbench	85.1	+0.1
livebench-instructions	85.1	+0.1
livebench-data	85.1	+0.1
analyst-agent	85.1	+0.1
livebench-math	84.9	-0.1
terminal-bench-science	85.1	+0.1
livebench-coding	85.0	-0.1
enigma-eval	85.1	+0.1
livebench-reasoning	85.0	-0.0
frontiermath	85.0	-0.0
livecodebench	85.0	-0.0
enterprise-ops	85.0	+0.0
swe-atlas-refactoring	85.0	-0.0
apex-agents	85.0	+0.0
gmmlu	85.0	-0.0
aime	85.0	+0.0
terminal-bench	85.0	-0.0
math500	85.0	-0.0
swe-bench-pro	85.0	+0.0

What evidence is missing from the fit?

Across all collected settings for this model: 20 observations, 17 contributing, 0 matched without graph weight and 3 excluded. Counts do not establish rank eligibility.

1 · Benchmark family or source protocol has not been reviewed
1 · Original SciCode grading defects are under review; retained as supporting evidence, excluded from aggregate and capability scores. Corrected SciCode-Verified requires a separate protocol review. Sources: https://arxiv.org/abs/2608.04975 and https://artificialanalysis.ai/evaluations/scicode
1 · Supporting evidence outside the reviewed capability core

Inspect every observation and exclusion →

Benchmark results

These are published benchmark scores. Supported effort estimates also appear on the Capability leaderboard. Results from different harnesses occupy different rows. Missing cells stay unknown; no scores are copied between effort levels.

Benchmark / source	Max
AA-Briefcase EloArtificial Analysis · Artificial Analysis published evaluation; see source for benchmark-specific settings Benchmark details AA-Briefcase v1.1 is an agentic knowledge work benchmark developed by Artificial Analysis. AA-Briefcase Elo is a combined metric that aggregates rubric pass rate, analytical quality Elo and presentation Elo · Higher is better · Evaluation results measured independently by Artificial Analysis Unit: elo. Source collected 2026-10-18.	1,729.32 Elo Reported settings Muse Wonder Full (Max); Model source: https://artificialanalysis.ai/models/muse-wonder-full
AA-LCR v1.1Artificial Analysis · Artificial Analysis published evaluation; see source for benchmark-specific settings Benchmark details Independently benchmarked by Artificial Analysis · Evaluation results measured independently by Artificial Analysis Unit: percent. Source collected 2026-10-18.	83.67% Reported settings Muse Wonder Full (Max); Model source: https://artificialanalysis.ai/models/muse-wonder-full
AutomationBench-AAArtificial Analysis · Artificial Analysis published evaluation; see source for benchmark-specific settings Benchmark details Share of task objectives completed with no guardrail violations · Higher is better · Benchmark developed by Zapier · Independently benchmarked by Artificial Analysis · Evaluation results measured independently by Artificial Analysis Unit: percent. Source collected 2026-10-18.	71.33% Reported settings Muse Wonder Full (Max); Model source: https://artificialanalysis.ai/models/muse-wonder-full
CritPtArtificial Analysis · Artificial Analysis published evaluation; see source for benchmark-specific settings Benchmark details Benchmark developed by Argonne and UIUC, with contributions from 60+ researchers globally. CritPt is under review by its authors following external feedback · Independently benchmarked by Artificial Analysis · Evaluation results measured independently by Artificial Analysis Unit: percent. Source collected 2026-10-18.	32.86% Reported settings Muse Wonder Full (Max); Model source: https://artificialanalysis.ai/models/muse-wonder-full
GDP.pdf: All-pass RateArtificial Analysis · Artificial Analysis published evaluation; see source for benchmark-specific settings Benchmark details Share of attempts where every atomic criterion passed · Independently benchmarked by Artificial Analysis · Evaluation results measured independently by Artificial Analysis Unit: percent. Source collected 2026-10-18.	32.6% Reported settings Muse Wonder Full (Max); Model source: https://artificialanalysis.ai/models/muse-wonder-full
GDPval-AA v2.1 LeaderboardArtificial Analysis · Artificial Analysis published evaluation; see source for benchmark-specific settings Benchmark details Elo rating for performance on real-world work tasks · Anchored to DeepSeek V4.1 Flash (max) at 1600 · Higher is better · Evaluation results measured independently by Artificial Analysis Unit: elo. Source collected 2026-10-18.	1,857.09 Elo Reported settings Muse Wonder Full (Max); Model source: https://artificialanalysis.ai/models/muse-wonder-full
Harvey LAB-AA v1.1: Hallucination-Gated All-Pass RateArtificial Analysis · Stirrup; Harvey private dataset v1.1.0; 120 tasks across 24 practice areas; up to 200 turns; internet disabled; GPT-6 Sol, Grok 4.7 and Claude Opus 5.5 rubric judges; GPT-6 Sol High two-pass hallucination check; hallucination-gated all-pass averaged across judges Benchmark details Average over tasks of the share of the judge panel finding every rubric criterion passed, with a task zeroed on any material hallucination · Independently benchmarked by Artificial Analysis · Evaluation results measured independently by Artificial Analysis Unit: percent. Source collected 2026-10-18.	13.06% Reported settings Muse Wonder Full (Max); Model source: https://artificialanalysis.ai/models/muse-wonder-full
Humanity's Last ExamArtificial Analysis · Artificial Analysis published evaluation; see source for benchmark-specific settings Benchmark details Independently benchmarked by Artificial Analysis · Evaluation results measured independently by Artificial Analysis Unit: percent. Source collected 2026-10-18.	53.61% Reported settings Muse Wonder Full (Max); Model source: https://artificialanalysis.ai/models/muse-wonder-full
AA-Omniscience IndexArtificial Analysis · Artificial Analysis published evaluation; see source for benchmark-specific settings Benchmark details AA-Omniscience Index (higher is better) measures knowledge reliability and hallucination. It rewards correct answers, penalizes hallucinations, and has no penalty for refusing to answer. Scores range from -100 to 100, where 0 means as many correct as incorrect answers, and negative scores mean more incorrect than correct. · Evaluation results measured independently by Artificial Analysis Unit: index. Source collected 2026-10-18.	30.15 Reported settings Muse Wonder Full (Max); Model source: https://artificialanalysis.ai/models/muse-wonder-full
AA-Omniscience AccuracyArtificial Analysis · 6000 questions; 1 attempt; correct-answer share; GPT-5.6 Luna medium grader Benchmark details AA-Omniscience Accuracy (higher is better) measures the proportion of correctly answered questions out of all questions, regardless of whether the model chooses to answer · Evaluation results measured independently by Artificial Analysis Unit: percent. Source collected 2026-10-18.	43.22% Reported settings Muse Wonder Full (Max); Model source: https://artificialanalysis.ai/models/muse-wonder-full
SciCodeArtificial Analysis · Artificial Analysis published evaluation; see source for benchmark-specific settings Benchmark details Benchmark curated by scientists across 16 disciplines. SciCode is under review following an independent audit of the dataset · Independently benchmarked by Artificial Analysis · Evaluation results measured independently by Artificial Analysis Unit: percent. Source collected 2026-10-18.	62.62% Reported settings Muse Wonder Full (Max); Model source: https://artificialanalysis.ai/models/muse-wonder-full
Terminal-Bench v4.0Artificial Analysis · mini-SWE-agent v2.4.6; 66 tasks; pass@1 averaged over 3 repeats; 500 steps; task timeout up to 8 hours Benchmark details Benchmark developed by the Laude Institute, Stanford, and open-source contributors · Independently benchmarked by Artificial Analysis · Evaluation results measured independently by Artificial Analysis Unit: percent. Source collected 2026-10-18.	57.07% Reported settings Muse Wonder Full (Max); Model source: https://artificialanalysis.ai/models/muse-wonder-full
Vals AI · Finance Agent v2Vals AI · Vals FAB v2 shared six-tool harness; two-hour task limit Benchmark details Overall weighted rubric accuracy, averaged over three runs on 450 tasks. Dealbreaker failures and timeouts score zero. Related categories and all-pass scores are not additional votes. Three-model judge panel; private test set. Unit: percent. Source collected 2026-10-18.	59.06% Reported settings {"checkpoint":"straw-110_evals_ext","evaluatorModel":"meta/muse-wonder-full","sourceUpdatedOn":"2026-10-16","reasoning_effort":"max","compute_effort":null,"reasoning":null,"temperature":1,"top_p":null,"max_output_tokens":131072,"verbosity":null,"provider":"Meta","harness":null}
Vals AI · Code Migration v1Vals AI · Vals Code Migration shared agent harness Benchmark details Hidden behavior-test pass rate, equally weighted across 40 repositories (30 CLI and 10 COBOL-to-Java). Anti-cheat failures score zero. Code-quality ratings and language splits are not additional votes. Unit: percent. Source collected 2026-10-18.	61.03% Reported settings {"checkpoint":"straw-110_evals_ext","evaluatorModel":"meta/muse-wonder-full","sourceUpdatedOn":"2026-10-16","reasoning_effort":"max","compute_effort":null,"reasoning":null,"temperature":1,"top_p":null,"max_output_tokens":131072,"verbosity":null,"provider":"Meta","harness":null}
Vals AI · Legal Research Bench v1Vals AI · Vals Legal Research shared five-tool harness; three-hour task limit Benchmark details All-pass accuracy on 208 private legal research tasks. Every expert rubric check must pass; GPT-5.4 judges responses. Partial-credit scores and practice-area splits are not additional votes. Unit: percent. Source collected 2026-10-18.	53.37% Reported settings {"checkpoint":"straw-110_evals_ext","evaluatorModel":"meta/muse-wonder-full","sourceUpdatedOn":"2026-10-16","reasoning_effort":"max","compute_effort":null,"reasoning":null,"temperature":1,"top_p":null,"max_output_tokens":131072,"verbosity":null,"provider":"Meta","harness":null}
Vals AI · Excel Modeling Benchmark v1Vals AI · Vals EMB shared workbook agent harness; 3.5-hour task limit Benchmark details Mean of seven equally weighted task categories. Template submissions use recalculated Excel cell accuracy; scratch submissions use rubric checks judged by GPT-5.4 Mini XHigh. Partial credit does not mean task completion. Unit: percent. Source collected 2026-10-18.	71.92% Reported settings {"checkpoint":"straw-110_evals_ext","evaluatorModel":"meta/muse-wonder-full","sourceUpdatedOn":"2026-10-16","reasoning_effort":"max","compute_effort":null,"reasoning":null,"temperature":1,"top_p":null,"max_output_tokens":131072,"verbosity":null,"provider":"Meta","harness":null}
Vals AI · Vibe Code Bench v1.1Vals AI · Vals modified OpenHands harness; five hours or 1,000 turns per app Benchmark details Mean app test pass rate on the canonical v1.1 test set. Browser Use executes workflow tests; each test requires at least 90% of its substeps. Automated judging can introduce error. Alternate agents are excluded. Unit: percent. Source collected 2026-10-18.	86% Reported settings {"checkpoint":"straw-110_evals_ext","evaluatorModel":"meta/muse-wonder-full","sourceUpdatedOn":"2026-10-16","reasoning_effort":"max","compute_effort":null,"reasoning":null,"temperature":1,"top_p":null,"max_output_tokens":131072,"verbosity":null,"provider":"Meta","harness":"OpenHands"}
Vals AI · Harvey Legal Agent Benchmark v1Vals AI · Vals Harvey held-out legal file-work protocol; internet disabled Benchmark details Mean task resolution across two judges (GPT-5.5 Medium and Sonnet 4.6), requiring every criterion to pass. Criteria pass rates and practice-area splits do not add votes. Shared Harvey family; Fable 5 fallback run excluded. Unit: percent. Source collected 2026-10-18.	21.67% Reported settings {"checkpoint":"straw-110_evals_ext","evaluatorModel":"meta/muse-wonder-full","sourceUpdatedOn":"2026-10-16","reasoning_effort":"max","compute_effort":null,"reasoning":null,"temperature":1,"top_p":null,"max_output_tokens":131072,"verbosity":null,"provider":"Meta","harness":null}
Vals AI · Tax Agent Bench v1: Weighted PassVals AI · Vals Tax Agent Bench shared five-tool harness; three-hour task limit Benchmark details Weighted Pass on 193 private US corporate-tax questions. Each task score is the weighted rubric share, set to zero when a must-pass check fails, then multiplied by (0.7 + 0.3 times citation quality). Timeouts score zero. Category splits and all-pass rates are not additional votes. GPT-5.4 judges expert rubrics. Unit: percent. Source collected 2026-10-18.	75% Reported settings {"checkpoint":"straw-110_evals_ext","evaluatorModel":"meta/muse-wonder-full","sourceUpdatedOn":"2026-10-16","reasoning_effort":"max","compute_effort":null,"reasoning":null,"temperature":1,"top_p":null,"max_output_tokens":131072,"verbosity":null,"provider":"Meta","harness":null}

What these comparisons mean

Each source reports its own effort labels. Matching labels do not establish equal compute budgets or identical fallback behavior. Open Reported settings to inspect the exact model name, agent and task coverage. Multiple results at one level remain visible rather than selecting the best. Source-reported None or Non-reasoning labels are not treated as evidence that an API supports those settings.

These measurements use their own dated source collection. The leaderboard fits identified configurations jointly; it contains no pooled model ratings. Explicit thinking budgets are separate configurations. Read the methodology.

Download all effort benchmark results · 8,915 results across 259 models · Collected 2026-10-18

Muse Wonder Full

Compare with

One configuration, seven capabilities

Agentic

Hard reasoning

Coding

Human pref

Knowledge

Multimodal

Long context

Capability score evidence

Benchmark results