UnifyBench

RankingEffort benchmarks

Muse Spark 1.3 · XHigh

Compare models

78 measured results · 4 effort settings · 63 benchmark/harness combinations

Model evidence and sources →

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 1.3 · 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 1.3 · XHigh · Agentic: 77.0 · SupportedMuse Spark 1.3 · XHigh · Hard reasoning: 66.6 · SupportedMuse Spark 1.3 · XHigh · Coding: 64.3 · SupportedMuse Spark 1.3 · XHigh · Knowledge: 52.5 · SupportedMuse Spark 1.3 · XHigh · Multimodal: 50.8 · PreliminaryMuse Spark 1.3 · XHigh · Long context: 79.7 · Supported

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

  • Agentic

    A77.0Supported

  • Hard reasoning

    A66.6Supported

  • Coding

    A64.3Supported

  • Human pref

    Unknown

  • Knowledge

    A52.5Supported

  • Multimodal

    A50.8Preliminary

  • Long context

    A79.7Supported

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 1.3 · XHigh: 74.3 overall comparison score. Meets the overall evidence requirements.

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

Shared-result disagreement. This aggregate order conflicts with matched benchmark results against 1 other configurations. Different test coverage and opponents affect the fit. Inspect published results before treating the rank as conclusive.

Which sources carry the weight?
  • Artificial Analysis · 31.6% direct comparison weight · independent of the model provider
  • Meta · 25.0% direct comparison weight · provider report
  • LiveBench · 22.2% direct comparison weight · independent of the model provider
  • Vals AI · 21.2% direct comparison weight · independent of the model provider
Which benchmark families carry the weight?
  • AutomationBench · 10.9% direct comparison weight
  • DeepSearchQA · 8.3% direct comparison weight
  • MRCR · 8.3% direct comparison weight
  • Terminal-Bench · 5.1% direct comparison weight
  • livebench-coding · 3.2% direct comparison weight
  • Other families (23) · 64.2% 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 sourceRank before → afterRank movementCohort remaining / lost supportScore after (change)
Artificial Analysis30 → 33Down 3203 remaining / 241 lost74.5 (+0.2 points)
ARC Prize30 → 33Down 3444 remaining / 0 lost76.9 (+2.7 points)
Datacurve30 → 31Down 1444 remaining / 0 lost71.7 (-2.6 points)
LiveCodeBench30 → 30No movement444 remaining / 0 lost74.3 (-0.0 points)
Scale30 → 31Down 1444 remaining / 0 lost75.0 (+0.8 points)
Terminal-Bench30 → 31Down 1444 remaining / 0 lost74.2 (-0.0 points)
Cognition30 → 28Up 2444 remaining / 0 lost73.9 (-0.4 points)
LiveBench30 → 39Down 9443 remaining / 1 lost73.3 (-0.9 points)
SWE-rebench30 → 30No movement444 remaining / 0 lost73.8 (-0.4 points)
Epoch AI30 → 36Down 6431 remaining / 13 lost73.9 (-0.4 points)
Vals AI30 → 27Up 3443 remaining / 1 lost75.6 (+1.3 points)
Provider reports · Anthropic30 → 32Down 2444 remaining / 0 lost74.2 (-0.1 points)
Provider reports · Moonshot AI30 → 30No movement444 remaining / 0 lost74.0 (-0.3 points)
Provider reports · Meta30 → 44Down 14444 remaining / 0 lost65.6 (-8.7 points)
Provider reports · Mistral30 → 30No movement444 remaining / 0 lost74.3 (+0.0 points)
Provider reports · OpenAI30 → 29Up 1444 remaining / 0 lost74.2 (-0.1 points)
All provider reports30 → 45Down 15444 remaining / 0 lost65.2 (-9.1 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
deepswe71.7-2.6
harvey72.4-1.8
vals-code-migration76.0+1.8
arc-agi75.9+1.6
livebench-language75.5+1.2
vals-finance-agent73.1-1.2
terminal-bench75.4+1.1
vals-excel-modeling75.2+1.0
multi-swe-bench73.3-0.9
livebench-instructions73.4-0.9
omniscience75.0+0.8
mmmu-pro74.9+0.7
vals-tax-agent73.7-0.6
swe-rebench73.8-0.4
frontiercode73.9-0.4
vibe-code74.0-0.3
tau-bench74.0-0.3
hle74.6+0.3
epoch-game-puzzles74.0-0.3
livebench-data74.5+0.3
ifbench74.0-0.3
gpqa74.0-0.2
gdpval74.0-0.2
swe-atlas-test-writing74.0-0.2
arc-agi-374.0-0.2
livebench-reasoning74.5+0.2
swe-atlas-qna74.1-0.2
mmlu-pro74.5+0.2
simpleqa74.1-0.2
briefcase74.1-0.1
livebench-coding74.4+0.1
vals-legal-research74.4+0.1
aime74.3+0.1
automationbench74.3+0.1
swe-atlas-refactoring74.2-0.0
enterprise-ops74.2-0.0
frontiermath74.3+0.0
enigma-eval74.3+0.0
apex-agents74.3+0.0
critpt74.3+0.0
livebench-math74.3-0.0
gdp-pdf74.3+0.0
terminal-bench-science74.3+0.0
swe-bench-pro74.3+0.0
livecodebench74.3-0.0
aa-lcr74.3-0.0
itbench74.3-0.0
math50074.3+0.0
gmmlu74.3+0.0
analyst-agent74.3+0.0
What evidence is missing from the fit?

Across all collected settings for this model: 114 observations, 65 contributing, 0 matched without graph weight and 49 excluded. Counts do not establish rank eligibility.

  • 4 · Benchmark family or source protocol has not been reviewed
  • 2 · 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 · Fewer than two distinct model identities remain in this evaluation
  • 9 · Evaluator has not been reviewed for independent admission
  • 1 · Effort is not identified
  • 2 · Secondary GDPval-AA v2 leaderboard report; the original evaluator already supplies this exact configuration. Reprinted or rerated board snapshots are not additional evaluation evidence.
  • 15 · Supporting evidence outside the reviewed capability core
  • 7 · No matched, identified effort unit; duplicates and unnamed comparators do not add evidence
  • 2 · No matched opponent in this evaluation unit
  • 2 · Composite index or publisher-specific internal evaluation lacks a stable general-capability comparison unit.
  • 2 · Different native coding harnesses are mixed; exact shared agent configuration is not established.
  • 2 · No reviewed effort for this catalog observation; any separately collected effort measurement is counted separately
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 / sourceHighXHighMaxUnspecified
AA-AnalystAgentArtificial Analysis · Artificial Analysis published evaluation; see source for benchmark-specific settings
Benchmark details

Share of tasks solved on all 5 of 5 attempts · 80 tasks, 5 attempts per task · Independently benchmarked by Artificial Analysis · Evaluation results measured independently by Artificial Analysis

Unit: percent. Source collected 2026-10-18.

Not reportedNot reported
37.5%
Reported settings

Muse Spark 1.3 (Max); Model source: https://artificialanalysis.ai/models/muse-spark-1-3

Not reported
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.

Not reported
1,485.51 Elo
Reported settings

Muse Spark 1.3 (Xhigh); Model source: https://artificialanalysis.ai/models/muse-spark-1-3-xhigh

1,581.34 Elo
Reported settings

Muse Spark 1.3 (Max); Model source: https://artificialanalysis.ai/models/muse-spark-1-3

Not reported
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.

Not reported
83%
Reported settings

Muse Spark 1.3 (Xhigh); Model source: https://artificialanalysis.ai/models/muse-spark-1-3-xhigh

83%
Reported settings

Muse Spark 1.3 (Max); Model source: https://artificialanalysis.ai/models/muse-spark-1-3

Not reported
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.

Not reported
56.84%
Reported settings

Muse Spark 1.3 (Xhigh); Model source: https://artificialanalysis.ai/models/muse-spark-1-3-xhigh

57.85%
Reported settings

Muse Spark 1.3 (Max); Model source: https://artificialanalysis.ai/models/muse-spark-1-3

Not reported
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.

Not reported
26%
Reported settings

Muse Spark 1.3 (Xhigh); Model source: https://artificialanalysis.ai/models/muse-spark-1-3-xhigh

24.86%
Reported settings

Muse Spark 1.3 (Max); Model source: https://artificialanalysis.ai/models/muse-spark-1-3

Not reported
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.

Not reported
24.2%
Reported settings

Muse Spark 1.3 (Xhigh); Model source: https://artificialanalysis.ai/models/muse-spark-1-3-xhigh

26.6%
Reported settings

Muse Spark 1.3 (Max); Model source: https://artificialanalysis.ai/models/muse-spark-1-3

Not reported
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.

Not reported
1,627.25 Elo
Reported settings

Muse Spark 1.3 (Xhigh); Model source: https://artificialanalysis.ai/models/muse-spark-1-3-xhigh

1,680.99 Elo
Reported settings

Muse Spark 1.3 (Max); Model source: https://artificialanalysis.ai/models/muse-spark-1-3

Not reported
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 2026-10-18.

Not reported
94.14%
Reported settings

Muse Spark 1.3 (Xhigh); Model source: https://artificialanalysis.ai/models/muse-spark-1-3-xhigh

93.54%
Reported settings

Muse Spark 1.3 (Max); Model source: https://artificialanalysis.ai/models/muse-spark-1-3

Not reported
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.

Not reportedNot reported
8.89%
Reported settings

Muse Spark 1.3 (Max); Model source: https://artificialanalysis.ai/models/muse-spark-1-3

Not reported
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.

Not reported
47.45%
Reported settings

Muse Spark 1.3 (Xhigh); Model source: https://artificialanalysis.ai/models/muse-spark-1-3-xhigh

48.7%
Reported settings

Muse Spark 1.3 (Max); Model source: https://artificialanalysis.ai/models/muse-spark-1-3

Not reported
ITBench-AAArtificial Analysis · Artificial Analysis published evaluation; see source for benchmark-specific settings
Benchmark details

Combined public and private Site Reliability Engineering (SRE) tasks · Independently benchmarked by Artificial Analysis · Evaluation results measured independently by Artificial Analysis

Unit: percent. Source collected 2026-10-18.

Not reportedNot reported
33.24%
Reported settings

Muse Spark 1.3 (Max); Model source: https://artificialanalysis.ai/models/muse-spark-1-3

Not reported
MLCR-AAArtificial Analysis · Artificial Analysis published evaluation; see source for benchmark-specific settings
Benchmark details

Share of tasks judged accurate, complete, and concise · Expert + compound (hard) sets · Benchmark developed by Wisedocs · Independently benchmarked by Artificial Analysis · Evaluation results measured independently by Artificial Analysis

Unit: percent. Source collected 2026-10-18.

Not reportedNot reported
43.33%
Reported settings

Muse Spark 1.3 (Max); Model source: https://artificialanalysis.ai/models/muse-spark-1-3

Not reported
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 2026-10-18.

Not reported
82.02%
Reported settings

Muse Spark 1.3 (Xhigh); Model source: https://artificialanalysis.ai/models/muse-spark-1-3-xhigh

Not reportedNot reported
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.

Not reported
23.12
Reported settings

Muse Spark 1.3 (Xhigh); Model source: https://artificialanalysis.ai/models/muse-spark-1-3-xhigh

25
Reported settings

Muse Spark 1.3 (Max); Model source: https://artificialanalysis.ai/models/muse-spark-1-3

Not reported
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.

Not reported
41.52%
Reported settings

Muse Spark 1.3 (Xhigh); Model source: https://artificialanalysis.ai/models/muse-spark-1-3-xhigh

43.58%
Reported settings

Muse Spark 1.3 (Max); Model source: https://artificialanalysis.ai/models/muse-spark-1-3

Not reported
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.

Not reported
59.72%
Reported settings

Muse Spark 1.3 (Xhigh); Model source: https://artificialanalysis.ai/models/muse-spark-1-3-xhigh

58.8%
Reported settings

Muse Spark 1.3 (Max); Model source: https://artificialanalysis.ai/models/muse-spark-1-3

Not reported
𝜏³-BankingArtificial Analysis · Artificial Analysis published evaluation; see source for benchmark-specific settings
Benchmark details

Benchmark developed by Sierra Research · Independently benchmarked by Artificial Analysis · Evaluation results measured independently by Artificial Analysis

Unit: percent. Source collected 2026-10-18.

Not reported
47.22%
Reported settings

Muse Spark 1.3 (Xhigh); Model source: https://artificialanalysis.ai/models/muse-spark-1-3-xhigh

50.52%
Reported settings

Muse Spark 1.3 (Max); Model source: https://artificialanalysis.ai/models/muse-spark-1-3

Not reported
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 2026-10-18.

Not reportedNot reported
10.95%
Reported settings

Muse Spark 1.3 (Max); Model source: https://artificialanalysis.ai/models/muse-spark-1-3

Not reported
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 2026-10-18.

Not reported
85.39%
Reported settings

Muse Spark 1.3 (Xhigh); Model source: https://artificialanalysis.ai/models/muse-spark-1-3-xhigh

84.27%
Reported settings

Muse Spark 1.3 (Max); Model source: https://artificialanalysis.ai/models/muse-spark-1-3

Not reported
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.

Not reported
16.67%
Reported settings

Muse Spark 1.3 (Xhigh); Model source: https://artificialanalysis.ai/models/muse-spark-1-3-xhigh

33.33%
Reported settings

Muse Spark 1.3 (Max); Model source: https://artificialanalysis.ai/models/muse-spark-1-3

Not reported
Terminal-Bench 4.0 · musecodeTerminal-Bench · Muse Code
Benchmark details

Terminal-Bench 4.0; Muse Code. Open each result for its exact effort, task subset, metric and protocol details.

Unit: percent. Source collected 2026-10-18.

Not reported
14.5%
Reported settings

Terminal-Bench 4.0.0; 66 tasks; Muse Code; effort xhigh; reported margin ±2.9 percentage points, confidence level unspecified; release date 2026-09-02; board checked 2026-10-07

Not reportedNot reported
LiveBench · AMPS Hard · 2026-06-25LiveBench · LiveBench official task evaluation
Benchmark details

LiveBench 2026-06-25 task snapshot. Task scores stay separate and related tasks share their reviewed family budget. Explicit checkpoint and effort labels only.

Unit: percent. Source collected 2026-10-18.

Not reported
99%
Reported settings

LiveBench official task evaluation; livebench-2026-06-25-AMPS_Hard; muse-spark-1.3-xhigh; checkpoint=muse-spark-1.3

Not reportedNot reported
LiveBench · code completion · 2026-06-25LiveBench · LiveBench official task evaluation
Benchmark details

LiveBench 2026-06-25 task snapshot. Task scores stay separate and related tasks share their reviewed family budget. Explicit checkpoint and effort labels only.

Unit: percent. Source collected 2026-10-18.

Not reported
80.44%
Reported settings

LiveBench official task evaluation; livebench-2026-06-25-code_completion; muse-spark-1.3-xhigh; checkpoint=muse-spark-1.3

Not reportedNot reported
LiveBench · code generation · 2026-06-25LiveBench · LiveBench official task evaluation
Benchmark details

LiveBench 2026-06-25 task snapshot. Task scores stay separate and related tasks share their reviewed family budget. Explicit checkpoint and effort labels only.

Unit: percent. Source collected 2026-10-18.

Not reported
81.69%
Reported settings

LiveBench official task evaluation; livebench-2026-06-25-code_generation; muse-spark-1.3-xhigh; checkpoint=muse-spark-1.3

Not reportedNot reported
LiveBench · connections · 2026-06-25LiveBench · LiveBench official task evaluation
Benchmark details

LiveBench 2026-06-25 task snapshot. Task scores stay separate and related tasks share their reviewed family budget. Explicit checkpoint and effort labels only.

Unit: percent. Source collected 2026-10-18.

Not reported
94%
Reported settings

LiveBench official task evaluation; livebench-2026-06-25-connections; muse-spark-1.3-xhigh; checkpoint=muse-spark-1.3

Not reportedNot reported
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