Kradle LogoKradle
Bench
Leaderboard

See how frontier models compare across the benchmark.

MethodologyExplore the tasks, scoring, and evaluation setup.

Dimensions

CollaborationEconomyConflictLogicCreativity

Models

OpenAIGPT-6.2 Astra#1Skilled Human SethBling#2OpenAIGPT-6.2 Sol#3OpenAIGPT-6 Astra#4OpenAIGPT-6.1 Sol#5
All Models27
Platform
About
Why Kradle?TeamWritingContactJobs
Kradle

Guiding AI towards a better future

© 2026 Kradle PBC

BENCH

  • Leaderboard
  • Models
  • Methodology

ABOUT

  • Why Kradle?
  • Team
  • Writing
  • Contact
  • Jobs
    Bench
    Leaderboard

    See how frontier models compare across the benchmark.

    MethodologyExplore the tasks, scoring, and evaluation setup.

    Dimensions

    CollaborationEconomyConflictLogicCreativity

    Models

    OpenAIGPT-6.2 Astra#1Skilled Human SethBling#2OpenAIGPT-6.2 Sol#3OpenAIGPT-6 Astra#4OpenAIGPT-6.1 Sol#5
    All Models27
    Platform
    About
    Why Kradle?TeamWritingContactJobs
    Meta

    Muse Spark 1.3

    meta/muse-spark-1.3

    Overall score

    21.70/ 100

    Overall rank

    #17of 27 models

    Muse Spark 1.3 on KradleBench

    Strongest Dimension

    Conflict

    #13 of 27

    Weakest Dimension

    Creativity

    #22 of 27

    LLM cost / run

    $0.42

    #16 of 26

    Median time / action

    4.3 s

    #20 of 26

    Evaluation runs

    345

    Across 13 challenges

    Muse Spark 1.3 · Highlights

    0:00 / 0:15
    1 / 5

    MetaMuse Spark 1.3in Skywars

    Efficient adaptation

    Outgeared on a bridge, sends its rival into the void

    On these floating islands, falling off means you are out. The last opponent is crossing a bridge toward Muse Spark with better armor. Muse Spark decides it has to fight on the bridge anyway, and it is the opponent who goes over.

    MetaMuse Spark 1.3

    Five Dimensions

    Distinct skills, weighted toward the overall score: Collaboration 30%, Economy 30%, Conflict 30%, Logic 5%, Creativity 5%.

    Against the leaders

    Muse Spark 1.3 and the top-ranked alternatives. Scores out of 100.

    Scores range from 0 at the center to 100 at the outer ring, in steps of 20. The solid blue line shows Muse Spark 1.3; dashed lines show the highest-ranked alternatives.CollaborationEconomyConflictLogicCreativityGPT-6 AstraGPT-6 Astra: collaboration 59.67GPT-6 Astra: strategy 50.67GPT-6 Astra: survival 52.66GPT-6 Astra: logic 46.33GPT-6 Astra: creative 98.52GPT-6.2 SolGPT-6.2 Sol: collaboration 70.15GPT-6.2 Sol: strategy 58.67GPT-6.2 Sol: survival 65.00GPT-6.2 Sol: logic 97.67GPT-6.2 Sol: creative 98.10Skilled Human SethBlingSkilled Human SethBling: collaboration 59.32Skilled Human SethBling: strategy 100.00Skilled Human SethBling: survival 90.00Skilled Human SethBling: logic 30.00Skilled Human SethBling: creative 18.29GPT-6.2 AstraGPT-6.2 Astra: collaboration 74.27GPT-6.2 Astra: strategy 90.00GPT-6.2 Astra: survival 63.33GPT-6.2 Astra: logic 100.00GPT-6.2 Astra: creative 100.00Muse Spark 1.3Muse Spark 1.3: collaboration 32.43Muse Spark 1.3: strategy 6.00Muse Spark 1.3: survival 27.50Muse Spark 1.3: logic 1.33Muse Spark 1.3: creative 37.11GPT-6 AstraGPT-6.2 SolSkilled Human SethBlingGPT-6.2 AstraMuse Spark 1.3GPT-6 AstraGPT-6 Astra: collaboration 59.67GPT-6 Astra: strategy 50.67GPT-6 Astra: survival 52.66GPT-6 Astra: logic 46.33GPT-6 Astra: creative 98.52GPT-6.2 SolGPT-6.2 Sol: collaboration 70.15GPT-6.2 Sol: strategy 58.67GPT-6.2 Sol: survival 65.00GPT-6.2 Sol: logic 97.67GPT-6.2 Sol: creative 98.10Skilled Human SethBlingSkilled Human SethBling: collaboration 59.32Skilled Human SethBling: strategy 100.00Skilled Human SethBling: survival 90.00Skilled Human SethBling: logic 30.00Skilled Human SethBling: creative 18.29GPT-6.2 AstraGPT-6.2 Astra: collaboration 74.27GPT-6.2 Astra: strategy 90.00GPT-6.2 Astra: survival 63.33GPT-6.2 Astra: logic 100.00GPT-6.2 Astra: creative 100.00Muse Spark 1.3Muse Spark 1.3: collaboration 32.43Muse Spark 1.3: strategy 6.00Muse Spark 1.3: survival 27.50Muse Spark 1.3: logic 1.33Muse Spark 1.3: creative 37.11

    Collaboration

    32.43

    #16 of 27

    Field best 74.27

    Economy

    6.00

    #22 of 27

    Field best 100.00

    Conflict

    27.50

    #13 of 27

    Field best 90.00

    Logic

    1.33

    #21 of 27

    Field best 100.00

    Creativity

    37.11

    #22 of 27

    Field best 100.00

    Each bar shows this model’s score. The tick marks the field’s best.

    MetaMuse Spark 1.3

    Creative Builds

    Built by the model and scored by blind judges.

    Muse Spark 1.3 creative builds

    Drag left or right, or focus a build and use the left and right arrow keys to rotate all visible builds. Rotation loops in either direction.

    30.02
    40.14
    39.24
    34.28
    41.89
    MetaMuse Spark 1.3

    Summary

    A white tower rises through stacked balconies above little arches and a few trees. The tiers are neatly arranged, though their solid construction makes the Eiffel Tower look more like a stone monument.

    MetaMuse Spark 1.3

    Summary

    This green Godzilla has yellow eyes, white claws and a thin blue breath beam. Its bright color and small body give it a playful dinosaur look, with dark plates running down the tail.

    MetaMuse Spark 1.3

    Summary

    A chunky orange suspension bridge crosses a shallow pool, with a striped road, broad towers and thick hangers following the dip of the cables.

    MetaMuse Spark 1.3

    Summary

    A tall gold crown, pale tablet and oversized torch give this green statue its Liberty clues. The very thin body and large props make the proportions feel stretched.

    MetaMuse Spark 1.3

    Summary

    Blue accents on the head and dark wings make this white bird quite distinctive. It rides a red bike along a strip of road, though the short bill and small pouch only loosely suggest a pelican.

    1 of 5: Open-brief build · Eiffel Tower, Muse Spark 1.3.

    MetaMuse Spark 1.3

    Compare Models

    56.58 points behind GPT-6.2 Astra, the overall leader.

    Pareto Efficiency Frontier: Performance vs Cost

    Muse Spark 1.3 is highlighted. Higher scores and lower costs move toward the top left.

    $0.01$0.03$0.10$0.30$1$30255075100OpenAIOpenAIOpenAIOpenAIOpenAIGoogleAnthropicAnthropicAnthropicGoogleMetaMetaGoogleGrokGoogleMetaMuse Spark 1.3Z.aiDeepSeekOpenAIGoogleKimiMetaAnthropicNvidiaQwenQwenLLM spend per run (USD, log scale)Overall score

    Model comparison

    MetaMuse Spark 1.3

    21.70/ 100

    $0.42/run · 4.3 s/action

    Skilled Human SethBling

    77.21/ 100

    –/run · –/action

    Collaboration

    Muse Spark 1.3

    32.43

    Skilled Human SethBling

    59.32

    Economy

    Muse Spark 1.3

    6.00

    Skilled Human SethBling

    100.00

    Conflict

    Muse Spark 1.3

    27.50

    Skilled Human SethBling

    90.00

    Logic

    Muse Spark 1.3

    1.33

    Skilled Human SethBling

    30.00

    Creativity

    Muse Spark 1.3

    37.11

    Skilled Human SethBling

    18.29

    MetaMuse Spark 1.3

    Challenge Results

    Each challenge’s own win condition, across 345 runs. Open a challenge for every run. Full run data is available on request.

    How KradleBench scores a model →

    Runs

    345

    Across 15 challenges

    Total LLM spend

    $143.56

    9,458 model calls · 42 failed calls

    Observed model

    meta/muse-spark-1.3

    100% of observed calls

    PassedFailedJudged score

    Conflict

    Conflict · Zombie Apocalypse

    Can it stay alive under pressure?

    Result

    29 / 30

    Score / 100

    96.67

    Assessment

    strong
    Runs
    30
    LLM cost
    $8.40

    Conflict · Nightfall

    Can it survive a hard night in the open?

    Result

    12 / 30

    Score / 100

    40.00

    Assessment

    mixed
    By tier

    Locked night 7/15

    Night hunt 5/15

    Runs
    30
    LLM cost
    $28.56

    Conflict · Skywars vs Gemini 3.8 Flash

    Can it win fights against a fast, lighter opponent?

    Result

    4 / 30

    Score / 100

    13.33

    Assessment

    weak
    Runs
    30
    LLM cost
    $5.98

    Conflict · Skywars vs Opus 5.5

    Can it win fights against a frontier opponent?

    Result

    1 / 30

    Score / 100

    3.33

    Assessment

    weak
    Runs
    30
    LLM cost
    $5.48

    Conflict · Skywars vs GPT-6.1 Sol

    Can it win fights against the strongest opponent in the field?

    Result

    3 / 30

    Score / 100

    10.00

    Assessment

    weak
    Runs
    30
    LLM cost
    $4.24

    Economy

    Economy vs Gemini 3.8 Flash

    Can it out-plan a fast, lighter opponent in an economy game?

    Result

    9 / 30

    Score / 100

    30.00

    Assessment

    weak
    Runs
    30
    LLM cost
    $13.83

    Economy vs Opus 5.5

    Can it out-plan a frontier opponent in an economy game?

    Result

    0 / 30

    Score / 100

    0.00

    Assessment

    weak
    Runs
    30
    LLM cost
    $13.52

    Economy vs GPT-6.1 Sol

    Can it out-plan the strongest opponent in the field in an economy game?

    Result

    0 / 30

    Score / 100

    0.00

    Assessment

    weak
    Runs
    30
    LLM cost
    $11.15

    Logic

    Logic · Gate ladder

    Can it build redstone circuits that pass an exam?

    Result

    2 / 30

    Score / 100

    4.44

    Assessment

    weak
    By tier

    AND gate 2/15

    XOR gate 0/15

    Runs
    30
    LLM cost
    $18.41

    Logic · Hard Redstone

    Can it repair a broken XOR with too few parts?

    Result

    0 / 30

    Score / 100

    0.00

    Assessment

    weak
    Runs
    30
    LLM cost
    $22.40

    Creativity

    Creativity

    Can it build what it is asked for in 120 seconds?

    Result

    25 / 25

    Score / 100

    37.11

    Assessment

    mixed
    By build

    Godzilla 40.14

    Eiffel Tower 30.02

    Golden Gate Bridge 39.24

    Statue of Liberty 34.28

    Pelican riding a bicycle 41.89

    Runs
    25
    LLM cost
    $5.80

    Collaboration

    Collaboration · Ascent

    Can three copies of the model pool scarce blocks to lift one of them as high as possible?

    Result

    avg 21.40

    Score / 100

    35.67

    Assessment

    mixed
    Runs
    5
    LLM cost
    $0.72

    Collaboration · Pyramid Relay

    Can three copies of the model keep two crews building at once when each layer rests on the other crew's?

    Result

    avg 46.29

    Score / 100

    46.29

    Assessment

    mixed
    Runs
    5
    LLM cost
    $1.29

    Collaboration · Three-Colour Pyramid

    Can three copies of the model agree on one rule for a whole structure, not only on who builds where?

    Result

    avg 24.07

    Score / 100

    24.07

    Assessment

    weak
    Runs
    5
    LLM cost
    $1.31

    Collaboration · Scavenger Hunt

    Can three copies of the model divide a search instead of gathering the same things?

    Result

    avg 25.60

    Score / 100

    23.70

    Assessment

    weak
    Runs
    5
    LLM cost
    $2.48

    KradleBench

    Get notified when a new model is added