RankingEffort benchmarks

Muse Spark Open

Compare models

24 measured results · 4 effort settings · 21 benchmark/harness combinations

Change model

Compare with

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

Compare with

Capability profile

One configuration, seven capabilities

Comparison score · 0–100
Muse Spark Open · XHigh capability radarComparison scores from 0 to 100. Unknown capabilities have no point; lines stop at gaps. Hollow points are preliminary. Exact values and support labels follow the chart.AgenticHard reasoningCodingHuman prefKnowledgeMultimodalLong context50100Muse Spark Open · XHigh · Agentic: 44.3 · SupportedMuse Spark Open · XHigh · Hard reasoning: 37.0 · SupportedMuse Spark Open · XHigh · Coding: 44.1 · PreliminaryMuse Spark Open · XHigh · Knowledge: 58.8 · SupportedMuse Spark Open · XHigh · Multimodal: 49.5 · PreliminaryMuse Spark Open · XHigh · Long context: 32.4 · Preliminary

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

  • Agentic

    A44.3Supported
  • Hard reasoning

    A37.0Supported
  • Coding

    A44.1Preliminary
  • Human pref

    Unknown
  • Knowledge

    A58.8Supported
  • Multimodal

    A49.5Preliminary
  • Long context

    A32.4Preliminary

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 Spark Open · XHigh: 33.4 overall comparison score. Meets the overall evidence requirements.

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

Which sources carry the weight?
  • Artificial Analysis · 36.0% direct comparison weight · independent of the model provider
  • ARC Prize · 35.8% direct comparison weight · independent of the model provider
  • Epoch AI · 28.1% direct comparison weight · independent of the model provider
Which benchmark families carry the weight?
  • arc-agi-3 · 18.5% direct comparison weight
  • ARC-AGI · 17.4% direct comparison weight
  • GPQA · 11.2% direct comparison weight
  • Aime · 8.0% direct comparison weight
  • simpleqa · 6.6% direct comparison weight
  • Other families (12) · 38.4% 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 2026-11-22.

Both ranks use the same surviving configurations for each removal. The cohort starts with 602 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 sourceRank before → afterRank movementCohort remaining / lost supportScore after (change)
Artificial Analysis118 → 120Down 2230 remaining / 372 lost31.3 (-2.1 points)
ARC Prize148 → 124Up 24602 remaining / 0 lost46.4 (+13.0 points)
Datacurve148 → 150Down 2602 remaining / 0 lost29.0 (-4.4 points)
LiveCodeBench148 → 148No movement602 remaining / 0 lost33.4 (-0.0 points)
Scale148 → 148No movement602 remaining / 0 lost35.3 (+1.8 points)
Terminal-Bench148 → 150Down 2602 remaining / 0 lost33.4 (-0.1 points)
Cognition148 → 154Down 6602 remaining / 0 lost32.0 (-1.4 points)
LiveBench148 → 146Up 2601 remaining / 1 lost34.6 (+1.1 points)
SWE-rebench148 → 149Down 1602 remaining / 0 lost33.3 (-0.1 points)
Epoch AI146 → 164Down 18588 remaining / 14 lost28.8 (-4.6 points)
Vals AI148 → 147Up 1601 remaining / 1 lost32.9 (-0.6 points)
Provider reports · Anthropic148 → 148No movement602 remaining / 0 lost33.4 (-0.0 points)
Provider reports · Moonshot AI148 → 149Down 1602 remaining / 0 lost33.0 (-0.5 points)
Provider reports · Meta148 → 148No movement602 remaining / 0 lost33.4 (-0.0 points)
Provider reports · Mistral148 → 148No movement602 remaining / 0 lost33.4 (+0.0 points)
Provider reports · OpenAI148 → 148No movement602 remaining / 0 lost33.6 (+0.1 points)
All provider reports148 → 149Down 1602 remaining / 0 lost33.1 (-0.4 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 evidenceRecomputed scoreChange
arc-agi-340.3+6.9
deepswe29.1-4.4
gpqa35.4+1.9
frontiercode32.0-1.4
mmmu-pro34.4+1.0
aime32.6-0.9
simpleqa32.6-0.8
terminal-bench-science34.1+0.7
aa-lcr34.1+0.6
hle34.0+0.6
omniscience32.9-0.5
epoch-game-puzzles32.9-0.5
mmlu-pro33.9+0.5
enigma-eval33.9+0.4
livebench-instructions33.8+0.4
multi-swe-bench33.1-0.4
briefcase33.1-0.4
automationbench33.1-0.3
arc-agi33.1-0.3
livebench-language33.7+0.3
swe-atlas-qna33.2-0.3
livebench-coding33.2-0.2
ifbench33.2-0.2
frontiermath33.6+0.2
vibe-code33.3-0.2
gdpval33.3-0.2
terminal-bench33.6+0.2
critpt33.6+0.2
tau-bench33.3-0.2
apex-agents33.6+0.2
livebench-reasoning33.3-0.1
swe-rebench33.3-0.1
gdp-pdf33.3-0.1
vals-legal-research33.3-0.1
vals-code-migration33.3-0.1
swe-atlas-test-writing33.3-0.1
vals-tax-agent33.4-0.1
vals-finance-agent33.4-0.1
swe-bench-pro33.5+0.1
itbench33.4-0.1
analyst-agent33.5+0.1
vals-excel-modeling33.4-0.1
livebench-data33.5+0.1
enterprise-ops33.4-0.1
swe-atlas-refactoring33.4-0.0
gmmlu33.5+0.0
math50033.5+0.0
livecodebench33.4-0.0
livebench-math33.4+0.0
harvey33.4-0.0
What evidence is missing from the fit?

Across all collected settings for this model: 24 observations, 24 contributing, 0 matched without graph weight and 0 excluded. Counts do not establish rank eligibility.

    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.

    Reviewed retained observations are in use. October 2026 Real-SWE board has no 95% confidence whisker, so these scores stay at the 2026-10-18 retrieval. Original source retrieval 2026-10-18; source updated 2026-10-18.

    Benchmark / sourceLowMediumHighXHigh
    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 2026-11-22.

    Not reportedNot reportedNot reported
    1,279.79 Elo
    Reported settings

    Muse Spark Open (Xhigh); Model source: https://artificialanalysis.ai/models/muse-spark-open

    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 2026-11-22.

    Not reportedNot reportedNot reported
    78.33%
    Reported settings

    Muse Spark Open (Xhigh); Model source: https://artificialanalysis.ai/models/muse-spark-open

    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 2026-11-22.

    Not reportedNot reportedNot reported
    49.71%
    Reported settings

    Muse Spark Open (Xhigh); Model source: https://artificialanalysis.ai/models/muse-spark-open

    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 · Independently benchmarked by Artificial Analysis · Evaluation results measured independently by Artificial Analysis

    Unit: percent. Source collected 2026-11-22.

    Not reportedNot reportedNot reported
    14.86%
    Reported settings

    Muse Spark Open (Xhigh); Model source: https://artificialanalysis.ai/models/muse-spark-open

    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 2026-11-22.

    Not reportedNot reportedNot reported
    18.6%
    Reported settings

    Muse Spark Open (Xhigh); Model source: https://artificialanalysis.ai/models/muse-spark-open

    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 2026-11-22.

    Not reportedNot reportedNot reported
    1,401.12 Elo
    Reported settings

    Muse Spark Open (Xhigh); Model source: https://artificialanalysis.ai/models/muse-spark-open

    GPQA DiamondArtificial 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 2026-11-22.

    Not reportedNot reportedNot reported
    88.99%
    Reported settings

    Muse Spark Open (Xhigh); Model source: https://artificialanalysis.ai/models/muse-spark-open

    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 2026-11-22.

    Not reportedNot reportedNot reported
    43.88%
    Reported settings

    Muse Spark Open (Xhigh); Model source: https://artificialanalysis.ai/models/muse-spark-open

    MLCR-AAArtificial Analysis · Artificial Analysis published evaluation; see source for benchmark-specific settings
    Benchmark details

    Share of responses judged complete with no claim contradicting the record · Expert + compound (hard) sets · Independently benchmarked by Artificial Analysis · Evaluation results measured independently by Artificial Analysis

    Unit: percent. Source collected 2026-11-22.

    Not reportedNot reportedNot reported
    7.78%
    Reported settings

    Muse Spark Open (Xhigh); Model source: https://artificialanalysis.ai/models/muse-spark-open

    MMMU-ProArtificial 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 2026-11-22.

    Not reportedNot reportedNot reported
    80.92%
    Reported settings

    Muse Spark Open (Xhigh); Model source: https://artificialanalysis.ai/models/muse-spark-open

    AA-Omniscience AccuracyArtificial Analysis · 3000 questions; 1 attempt; correct-answer share; GPT-5.6 Luna medium grader
    Benchmark details

    AA-Omniscience v1.1 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 2026-11-22.

    Not reportedNot reportedNot reported
    42.03%
    Reported settings

    Muse Spark Open (Xhigh); Model source: https://artificialanalysis.ai/models/muse-spark-open

    Terminal-Bench-Science 0.1Artificial Analysis · mini-swe-agent; Terminal-Bench-Science 0.1.0; 70 tasks; pass@1 averaged over 3 repeats
    Benchmark details

    Share of tasks passed (pass@1), averaged over three repeats · Independently benchmarked by Artificial Analysis · Evaluation results measured independently by Artificial Analysis

    Unit: percent. Source collected 2026-11-22.

    Not reportedNot reportedNot reported
    1.9%
    Reported settings

    Muse Spark Open (Xhigh); Model source: https://artificialanalysis.ai/models/muse-spark-open

    Terminal-Bench v2.1Artificial Analysis · Artificial Analysis published evaluation; see source for benchmark-specific settings
    Benchmark details

    Benchmark developed by the Laude Institute and open-source contributors · Independently benchmarked by Artificial Analysis · Evaluation results measured independently by Artificial Analysis

    Unit: percent. Source collected 2026-11-22.

    Not reportedNot reportedNot reported
    76.4%
    Reported settings

    Muse Spark Open (Xhigh); Model source: https://artificialanalysis.ai/models/muse-spark-open

    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 2026-11-22.

    Not reportedNot reportedNot reported
    8.08%
    Reported settings

    Muse Spark Open (Xhigh); Model source: https://artificialanalysis.ai/models/muse-spark-open

    ARC-AGI-2ARC Prize · v2 Semi-Private
    Benchmark details

    Original public board. Configuration and task coverage may differ; inspect source.

    Unit: percent. Source collected 2026-11-22.

    12.22%
    Reported settings

    Muse Spark Open (Low); Source model ID: meta-muse-spark-open-low; 120 tasks

    27.5%
    Reported settings

    Muse Spark Open (Medium); Source model ID: meta-muse-spark-open-medium; 120 tasks

    47.5%
    Reported settings

    Muse Spark Open (High); Source model ID: meta-muse-spark-open-high; 120 tasks

    57.5%
    Reported settings

    Muse Spark Open (XHigh); Source model ID: meta-muse-spark-open-xhigh; 120 tasks

    ARC-AGI-3 · provider-adapterARC Prize · Provider Adapter
    Benchmark details

    ARC-AGI-3; Provider Adapter. Open each result for its exact effort, task subset, metric and protocol details.

    Unit: percent. Source collected 2026-11-22.

    Not reportedNot reportedNot reported
    0.42%
    Reported settings

    ARC-AGI-3 Semi-Private verified summary; Provider Adapter harness; source group meta-muse-spark-open; variant provider-adapter-xhigh; public-demo task scores are excluded

    ARC-AGI-3 · standardARC Prize · Standard
    Benchmark details

    ARC-AGI-3; Standard. Open each result for its exact effort, task subset, metric and protocol details.

    Unit: percent. Source collected 2026-11-22.

    Not reportedNot reportedNot reported
    0.14%
    Reported settings

    ARC-AGI-3 Semi-Private verified summary; Standard harness; source group meta-muse-spark-open; variant xhigh; public-demo task scores are excluded

    Epoch · FrontierMath Tiers 1–3 · v2.0.0Epoch AI · Epoch AI Inspect; task 2.0.0; scorer verification_code
    Benchmark details

    Epoch AI internal evaluation. Mean accuracy under the published setting; exact task version and scorer stay separate. Related tasks share a family budget. Data by Epoch AI, CC BY. FrontierMath was developed with OpenAI funding and unequal access to part of the question set; this result is from Epoch's evaluation.

    Unit: percent. Source collected 2026-11-22.

    Not reportedNot reportedNot reported
    69.82%
    Reported settings

    Epoch AI Inspect; task 2.0.0; scorer verification_code; model=Muse-Spark-Open-816B-A42B_xhigh; run=e3e3h9r7Fouu4wsAeVzTnV; mean accuracy; https://epoch.ai/data/benchmarks.csv

    Epoch · GPQA diamond · v1.0.11Epoch AI · Epoch AI Inspect; task 1.0.11; scorer choice
    Benchmark details

    Epoch AI internal evaluation. Mean accuracy under the published setting; exact task version and scorer stay separate. Related tasks share a family budget. Data by Epoch AI, CC BY.

    Unit: percent. Source collected 2026-11-22.

    Not reportedNot reportedNot reported
    88.57%
    Reported settings

    Epoch AI Inspect; task 1.0.11; scorer choice; model=Muse-Spark-Open-816B-A42B_xhigh; run=JnKpor8c3EKWGvmyjLFVzH; mean accuracy; https://epoch.ai/data/benchmarks.csv

    Epoch · OTIS Mock AIME 2024-2025 · v1.1.6Epoch AI · Epoch AI Inspect; task 1.1.6; scorer model_extracted_exact_match
    Benchmark details

    Epoch AI internal evaluation. Mean accuracy under the published setting; exact task version and scorer stay separate. Related tasks share a family budget. Data by Epoch AI, CC BY.

    Unit: percent. Source collected 2026-11-22.

    Not reportedNot reportedNot reported
    98.06%
    Reported settings

    Epoch AI Inspect; task 1.1.6; scorer model_extracted_exact_match; model=Muse-Spark-Open-816B-A42B_xhigh; run=hB4XzBdWkdpG3WJVYqeRM8; mean accuracy; https://epoch.ai/data/benchmarks.csv

    Epoch · SimpleQA Verified · v1.2.0Epoch AI · Epoch AI Inspect; task 1.2.0; scorer simpleqa_scorer
    Benchmark details

    Epoch AI internal evaluation. Mean accuracy under the published setting; exact task version and scorer stay separate. Related tasks share a family budget. Data by Epoch AI, CC BY.

    Unit: percent. Source collected 2026-11-22.

    Not reportedNot reportedNot reported
    56.6%
    Reported settings

    Epoch AI Inspect; task 1.2.0; scorer simpleqa_scorer; model=Muse-Spark-Open-816B-A42B_xhigh; run=dNuxLdV4DWqfSvtTq6NnK8; mean accuracy; https://epoch.ai/data/benchmarks.csv

    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 · 10,184 results across 381 models · Collected 2026-11-22