GeekRadar / The comparison desk

COMPARE.

AI models

Muse Spark

AI model Released Apr 8, 2026

Capability & preference

General capability · ECI

A cross-benchmark capability estimate. Index points, not a percentage; intervals show uncertainty.

Muse Spark 152.2^* Muse Spark · 95% interval 150.0–154.4

Human preference · Arena

Overall text preferences from head-to-head votes. Not factual accuracy or a task-success rate.

Muse Spark 1474.0 Muse Spark · 95% interval 1468.3–1479.7 · 14,140 votes · 2026-10-28

Model performance

Radar task score: four tests, equal weight. An experimental summary of these tasks, not a universal intelligence rating.

Muse Spark

Tested configuration Radar task score

Unrated

1/4 core tests · not enough evidence

2 benchmark results for this configuration

Scientific reasoning

GPQA Diamond · 25%

Expert-written science questions; strict multiple-choice scoring. Chance accuracy is 25%.

Muse Spark 89.8^*% Approx. 95% interval: 85.5–94.1% · 2026-04-08 · Older result

Novel problem solving

Mystery Game Puzzles · 25%

Best-next-move puzzles in an undisclosed game variant. A narrow reasoning test, not general intelligence.

Muse Spark Not available

Advanced mathematics

FrontierMath Tiers 1-3 · 25%

Private v2 mathematics problems with Python tools. Earlier versions and Tier 4 are excluded.

Muse Spark Not available

Factual answers

SimpleQA Verified · 25%

Correct factual answers under a guessing prompt, graded by a model. Not a hallucination or abstention score.

Muse Spark Not available

Competition mathematics

OTIS Mock AIME 2024-2025 · Separate result

Competition problems with model-extracted answers checked against an exact key. Earlier grading procedures are excluded.

Muse Spark 88.9^*% Approx. 95% interval: 79.7–98.1% · 2026-04-08 · Older result

Bars show reported task scores. Whiskers are approximate 95% intervals where standard errors are available. Different reasoning budgets are not equal-cost tests. Missing results are never zero.

Recorded results

Only the same benchmark, version, workload, scenario and unit share a row. Missing results are not zero. System results include the full submitted hardware and software configuration.

Arena text preference / chinesehuman preference · 2026-10-28 · rating

Muse Spark

1,507.35 rating

Confidence interval: 1,487.57 – 1,527.14

976 votes

Oct 28, 2026

Category: chinese
Evaluated configuration: Muse Spark
Evaluation type: human preference

Arena text preference / codinghuman preference · 2026-10-28 · rating

Muse Spark

1,481.31 rating

Confidence interval: 1,471.46 – 1,491.15

3941 votes

Oct 28, 2026

Category: coding
Evaluated configuration: Muse Spark
Evaluation type: human preference

Arena text preference / creative_writinghuman preference · 2026-10-28 · rating

Muse Spark

1,458.71 rating

Confidence interval: 1,445.12 – 1,472.31

2038 votes

Oct 28, 2026

Category: creative_writing
Evaluated configuration: Muse Spark
Evaluation type: human preference

Arena text preference / englishhuman preference · 2026-10-28 · rating

Muse Spark

1,480.28 rating

Confidence interval: 1,472.65 – 1,487.92

6679 votes

Oct 28, 2026

Category: english
Evaluated configuration: Muse Spark
Evaluation type: human preference

Arena text preference / exclude_tieshuman preference · 2026-10-28 · rating

Muse Spark

1,479.32 rating

Confidence interval: 1,471.90 – 1,486.74

10714 votes

Oct 28, 2026

Category: exclude_ties
Evaluated configuration: Muse Spark
Evaluation type: human preference

Arena text preference / experthuman preference · 2026-10-28 · rating

Muse Spark

1,457.14 rating

Confidence interval: 1,441.16 – 1,473.12

1377 votes

Oct 28, 2026

Category: expert
Evaluated configuration: Muse Spark
Evaluation type: human preference

Arena text preference / frenchhuman preference · 2026-10-28 · rating

Muse Spark

1,495.58 rating

Confidence interval: 1,467.53 – 1,523.64

478 votes

Oct 28, 2026

Category: french
Evaluated configuration: Muse Spark
Evaluation type: human preference

Arena text preference / germanhuman preference · 2026-10-28 · rating

Muse Spark

1,498.05 rating

Confidence interval: 1,459.06 – 1,537.03

236 votes

Oct 28, 2026

Category: german
Evaluated configuration: Muse Spark
Evaluation type: human preference

Arena text preference / hard_promptshuman preference · 2026-10-28 · rating

Muse Spark

1,474.20 rating

Confidence interval: 1,467.38 – 1,481.03

9155 votes

Oct 28, 2026

Category: hard_prompts
Evaluated configuration: Muse Spark
Evaluation type: human preference

Arena text preference / hard_prompts_englishhuman preference · 2026-10-28 · rating

Muse Spark

1,477.24 rating

Confidence interval: 1,468.03 – 1,486.44

4506 votes

Oct 28, 2026

Category: hard_prompts_english
Evaluated configuration: Muse Spark
Evaluation type: human preference

Arena text preference / industry_business_and_management_and_financial_operationshuman preference · 2026-10-28 · rating

Muse Spark

1,462.04 rating

Confidence interval: 1,450.63 – 1,473.45

2863 votes

Oct 28, 2026

Category: industry_business_and_management_and_financial_operations
Evaluated configuration: Muse Spark
Evaluation type: human preference

Arena text preference / industry_entertainment_and_sports_and_mediahuman preference · 2026-10-28 · rating

Muse Spark

1,448.64 rating

Confidence interval: 1,436.69 – 1,460.60

2634 votes

Oct 28, 2026

Category: industry_entertainment_and_sports_and_media
Evaluated configuration: Muse Spark
Evaluation type: human preference

Arena text preference / industry_legal_and_governmenthuman preference · 2026-10-28 · rating

Muse Spark

1,484.20 rating

Confidence interval: 1,465.18 – 1,503.22

1019 votes

Oct 28, 2026

Category: industry_legal_and_government
Evaluated configuration: Muse Spark
Evaluation type: human preference

Arena text preference / industry_life_and_physical_and_social_sciencehuman preference · 2026-10-28 · rating

Muse Spark

1,475.45 rating

Confidence interval: 1,463.01 – 1,487.88

2372 votes

Oct 28, 2026

Category: industry_life_and_physical_and_social_science
Evaluated configuration: Muse Spark
Evaluation type: human preference

Arena text preference / industry_mathematicalhuman preference · 2026-10-28 · rating

Muse Spark

1,454.67 rating

Confidence interval: 1,434.44 – 1,474.90

859 votes

Oct 28, 2026

Category: industry_mathematical
Evaluated configuration: Muse Spark
Evaluation type: human preference

Arena text preference / industry_medicine_and_healthcarehuman preference · 2026-10-28 · rating

Muse Spark

1,475.47 rating

Confidence interval: 1,456.33 – 1,494.62

994 votes

Oct 28, 2026

Category: industry_medicine_and_healthcare
Evaluated configuration: Muse Spark
Evaluation type: human preference

Arena text preference / industry_software_and_it_serviceshuman preference · 2026-10-28 · rating

Muse Spark

1,486.85 rating

Confidence interval: 1,478.52 – 1,495.18

5551 votes

Oct 28, 2026

Category: industry_software_and_it_services
Evaluated configuration: Muse Spark
Evaluation type: human preference

Arena text preference / industry_writing_and_literature_and_languagehuman preference · 2026-10-28 · rating

Muse Spark

1,453.81 rating

Confidence interval: 1,443.16 – 1,464.46

3295 votes

Oct 28, 2026

Category: industry_writing_and_literature_and_language
Evaluated configuration: Muse Spark
Evaluation type: human preference

Arena text preference / instruction_followinghuman preference · 2026-10-28 · rating

Muse Spark

1,442.31 rating

Confidence interval: 1,433.36 – 1,451.26

4739 votes

Oct 28, 2026

Category: instruction_following
Evaluated configuration: Muse Spark
Evaluation type: human preference

Arena text preference / koreanhuman preference · 2026-10-28 · rating

Muse Spark

1,458.72 rating

Confidence interval: 1,420.05 – 1,497.40

251 votes

Oct 28, 2026

Category: korean
Evaluated configuration: Muse Spark
Evaluation type: human preference

Arena text preference / longer_queryhuman preference · 2026-10-28 · rating

Muse Spark

1,450.34 rating

Confidence interval: 1,441.82 – 1,458.87

5602 votes

Oct 28, 2026

Category: longer_query
Evaluated configuration: Muse Spark
Evaluation type: human preference

Arena text preference / mathhuman preference · 2026-10-28 · rating

Muse Spark

1,455.79 rating

Confidence interval: 1,436.32 – 1,475.27

892 votes

Oct 28, 2026

Category: math
Evaluated configuration: Muse Spark
Evaluation type: human preference

Arena text preference / multi_turnhuman preference · 2026-10-28 · rating

Muse Spark

1,476.70 rating

Confidence interval: 1,463.62 – 1,489.78

2282 votes

Oct 28, 2026

Category: multi_turn
Evaluated configuration: Muse Spark
Evaluation type: human preference

Arena text preference / non_englishhuman preference · 2026-10-28 · rating

Muse Spark

1,463.34 rating

Confidence interval: 1,455.99 – 1,470.69

7461 votes

Oct 28, 2026

Category: non_english
Evaluated configuration: Muse Spark
Evaluation type: human preference

Arena text preference / overallhuman preference · 2026-10-28 · rating

Muse Spark

1,474.04 rating

Confidence interval: 1,468.33 – 1,479.74

14140 votes

Oct 28, 2026

Category: overall
Evaluated configuration: Muse Spark
Evaluation type: human preference

Arena text preference / polishhuman preference · 2026-10-28 · rating

Muse Spark

1,474.55 rating

Confidence interval: 1,441.72 – 1,507.39

312 votes

Oct 28, 2026

Category: polish
Evaluated configuration: Muse Spark
Evaluation type: human preference

Arena text preference / russianhuman preference · 2026-10-28 · rating

Muse Spark

1,466.36 rating

Confidence interval: 1,451.22 – 1,481.49

1574 votes

Oct 28, 2026

Category: russian
Evaluated configuration: Muse Spark
Evaluation type: human preference

Arena text preference / spanishhuman preference · 2026-10-28 · rating

Muse Spark

1,469.56 rating

Confidence interval: 1,443.62 – 1,495.51

519 votes

Oct 28, 2026

Category: spanish
Evaluated configuration: Muse Spark
Evaluation type: human preference

Run it locally

Inference · memory estimate

Muse Spark

Not enough verified architecture information for a reliable VRAM estimate.

Capabilities

Model license: Proprietary

* Provisional data