BenchAtlas

Rankings / Meta AI

Muse Spark

released 2026-04-08

BenchAtlas Index

as of 2026-07-14
71.3
base configuration
rank #60
11 families · 5 categories · high
64.8
base configuration
rank #97
8 families · 4 categories · medium

Benchmark evidence

48 results
agentic coding
τ²-Bench91.5 %independentArtificial Analysis
τ²-Bench
subset=banking
19.6 %independentArtificial Analysis
coding
SciCode51.5 %independentArtificial Analysis
external indices
AA Coding Index58.6 pointsindependentArtificial Analysis
Intelligence Index v4.143.1 pointsindependentArtificial Analysis
ECI154.3 pointsindependentEpoch AI Benchmarking Hub
factuality
SimpleQA Verified66.3 %independentEpoch AI Benchmarking Hub
2026-04-08
human preference
Coding (style control)older version
arena=text · category=coding · style_control=true
1525.2
1515.11535.4
communityLMArena Leaderboard Dataset
Chinese (style control)older version
arena=text · category=chinese · style_control=true
1523.3
1501.71544.9
communityLMArena Leaderboard Dataset
Industry Software And It Services (style control)older version
arena=text · category=industry_software_and_it_services · style_control=true
1518.9
1510.31527.5
communityLMArena Leaderboard Dataset
Industry Medicine And Healthcare (style control)older version
arena=text · category=industry_medicine_and_healthcare · style_control=true
1506.9
1487.11526.8
communityLMArena Leaderboard Dataset
Hard Prompts English (style control)older version
arena=text · category=hard_prompts_english · style_control=true
1506.1
1496.61515.6
communityLMArena Leaderboard Dataset
Industry Legal And Government (style control)older version
arena=text · category=industry_legal_and_government · style_control=true
1504.8
1484.91524.7
communityLMArena Leaderboard Dataset
Hard Prompts (style control)older version
arena=text · category=hard_prompts · style_control=true
1504.3
1497.31511.4
communityLMArena Leaderboard Dataset
Exclude Ties (style control)older version
arena=text · category=exclude_ties · style_control=true
1500.0
1492.41507.7
communityLMArena Leaderboard Dataset
Industry Life And Physical And Social Science (style control)older version
arena=text · category=industry_life_and_physical_and_social_science · style_control=true
1498.1
1485.21511.0
communityLMArena Leaderboard Dataset
English (style control)older version
arena=text · category=english · style_control=true
1494.9
1487.11502.7
communityLMArena Leaderboard Dataset
Multi Turn (style control)older version
arena=text · category=multi_turn · style_control=true
1492.3
1478.71505.8
communityLMArena Leaderboard Dataset
Industry Business And Management And Financial Operations (style control)older version
arena=text · category=industry_business_and_management_and_financial_operations · style_control=true
1488.6
1476.71500.5
communityLMArena Leaderboard Dataset
Overall (style control)
arena=text · category=overall · style_control=true
1487.6
1481.71493.5
communityLMArena Leaderboard Dataset
Expert (style control)older version
arena=text · category=expert · style_control=true
1486.4
1469.41503.3
communityLMArena Leaderboard Dataset
Russian (style control)older version
arena=text · category=russian · style_control=true
1481.5
1465.61497.4
communityLMArena Leaderboard Dataset
Non English (style control)older version
arena=text · category=non_english · style_control=true
1477.0
1469.41484.6
communityLMArena Leaderboard Dataset
Longer Query (style control)older version
arena=text · category=longer_query · style_control=true
1473.5
1464.51482.4
communityLMArena Leaderboard Dataset
Industry Mathematical (style control)older version
arena=text · category=industry_mathematical · style_control=true
1469.5
1448.51490.4
communityLMArena Leaderboard Dataset
Math (style control)older version
arena=text · category=math · style_control=true
1463.1
1443.21483.1
communityLMArena Leaderboard Dataset
Instruction Following (style control)older version
arena=text · category=instruction_following · style_control=true
1463.0
1453.71472.3
communityLMArena Leaderboard Dataset
Creative Writing (style control)older version
arena=text · category=creative_writing · style_control=true
1462.4
1448.31476.5
communityLMArena Leaderboard Dataset
Industry Entertainment And Sports And Media (style control)older version
arena=text · category=industry_entertainment_and_sports_and_media · style_control=true
1460.5
1448.21472.9
communityLMArena Leaderboard Dataset
Industry Writing And Literature And Language (style control)older version
arena=text · category=industry_writing_and_literature_and_language · style_control=true
1458.1
1447.01469.1
communityLMArena Leaderboard Dataset
knowledge science
GPQA Diamond89.8 %independentEpoch AI Benchmarking Hub
2026-04-08
GPQA Diamond
implementation=vals-ai
89.6 %independentVals AI
GPQA Diamond
implementation=artificial-analysis
88.4 %independentArtificial Analysis
Humanity's Last Exam
contamination=Potential contamination warning: This model was evaluated after the public release of HLE, allowing model builder access to the prompts and solutions. · implementation=scale
40.6 %
38.642.5
independentScale Labs
Humanity's Last Exam
implementation=artificial-analysis
39.9 %independentArtificial Analysis
MMLU-Pro
implementation=vals-ai
87.3 %independentVals AI
long context instruction
AA-LCR69.7 %independentArtificial Analysis
IFBench75.9 %independentArtificial Analysis
MultiChallenge75.5 %
71.579.6
independentScale Labs
multimodal
MMMU
implementation=vals-ai
87.4 %independentVals AI
professional
CaseLaw63.1 %independentVals AI
CorpFin65.1 %independentVals AI
LegalBench84.2 %independentVals AI
TaxEval77.7 %independentVals AI
reasoning math
AIME
implementation=vals-ai
96.9 %independentVals AI
FrontierMath39.0 %independentEpoch AI Benchmarking Hub
2026-04-08
FrontierMath Tier 4older version14.6 %independentEpoch AI Benchmarking Hub
2026-04-08
OTIS Mock AIME 2024–202588.9 %independentEpoch AI Benchmarking Hub
2026-04-08

Agent + model results

systems, not bare-model scores
agent + model Artificial Analysis harness + Muse SparkTerminal-Bench 2.162.2 %independentArtificial Analysis
agent + model Artificial Analysis harness + Muse SparkTerminal-Bench Hard45.5 %independentArtificial Analysis

These scores measure the whole agent system (scaffold, tools, budgets) — they are never merged into the bare model’s numbers.

Related variants