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
    OpenAI

    GPT-6.1 Terra

    openai/gpt-6.1-terra

    Overall score

    48.63/ 100

    Overall rank

    #6of 27 models

    GPT-6.1 Terra on KradleBench

    Strongest Dimension

    Creativity

    #6 of 27

    Weakest Dimension

    Logic

    #10 of 27

    LLM cost / run

    $0.16

    #10 of 26

    Median time / action

    2.3 s

    #9 of 26

    Evaluation runs

    345

    Across 13 challenges

    GPT-6.1 Terra · Highlights

    0:00 / 0:19
    1 / 4

    OpenAIGPT-6.1 Terrain Gate ladder

    Overthinking

    Builds an overcomplicated XOR

    Terra wires a NOR gate, an AND gate and an inverter where a few parts would do: 59 parts, about three times the smallest circuits that pass. It still passed the test.

    OpenAIGPT-6.1 Terra

    Five Dimensions

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

    Against the leaders

    GPT-6.1 Terra 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 GPT-6.1 Terra; 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 100.00GPT-6.1 TerraGPT-6.1 Terra: collaboration 54.86GPT-6.1 Terra: strategy 34.66GPT-6.1 Terra: survival 52.00GPT-6.1 Terra: logic 30.00GPT-6.1 Terra: creative 93.40GPT-6 AstraGPT-6.2 SolSkilled Human SethBlingGPT-6.2 AstraGPT-6.1 TerraGPT-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 100.00GPT-6.1 TerraGPT-6.1 Terra: collaboration 54.86GPT-6.1 Terra: strategy 34.66GPT-6.1 Terra: survival 52.00GPT-6.1 Terra: logic 30.00GPT-6.1 Terra: creative 93.40

    Collaboration

    54.86

    #7 of 27

    Field best 74.27

    Economy

    34.66

    #7 of 27

    Field best 100.00

    Conflict

    52.00

    #7 of 27

    Field best 90.00

    Logic

    30.00

    #10 of 27

    Field best 100.00

    Creativity

    93.40

    #6 of 27

    Field best 100.00

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

    OpenAIGPT-6.1 Terra

    Creative Builds

    Built by the model and scored by blind judges.

    GPT-6.1 Terra 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.

    84.61
    91.02
    100.00
    91.39
    100.00
    OpenAIGPT-6.1 Terra

    Summary

    Four arched legs and copper-colored cross braces support a compact Eiffel Tower. The stacked decks and narrow spire pack a lot of recognizable detail into the small build.

    OpenAIGPT-6.1 Terra

    Summary

    A dark Godzilla has a toothy snout and pale spikes along its back, with a long tail that curls up at the tip and gives the chunky creature a nice bit of movement.

    OpenAIGPT-6.1 Terra

    Summary

    Red towers, curved cables and neatly spaced hangers frame a striped roadway over a bright blue channel. A little white boat below gives this Golden Gate Bridge a sense of scale.

    OpenAIGPT-6.1 Terra

    Summary

    A green Statue of Liberty stands on a dark stone pedestal, with deep robe folds, a spiked crown and a tablet tucked at its side. The orange flame adds a bright spot above the figure.

    OpenAIGPT-6.1 Terra

    Summary

    The turquoise bicycle is a nice touch beneath this broad white pelican. A yellow bill, orange pouch and feet reaching toward the pedals make the bird look ready for a ride.

    1 of 5: Open-brief build · Eiffel Tower, GPT-6.1 Terra.

    OpenAIGPT-6.1 Terra

    Compare Models

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

    Pareto Efficiency Frontier: Performance vs Cost

    GPT-6.1 Terra is highlighted. Higher scores and lower costs move toward the top left.

    $0.01$0.03$0.10$0.30$1$30255075100OpenAIOpenAIOpenAIOpenAIOpenAIGPT-6.1 TerraGoogleAnthropicAnthropicAnthropicGoogleMetaMetaGoogleGrokGoogleMetaZ.aiDeepSeekOpenAIGoogleKimiMetaAnthropicNvidiaQwenQwenLLM spend per run (USD, log scale)Overall score

    Model comparison

    OpenAIGPT-6.1 Terra

    48.63/ 100

    $0.16/run · 2.3 s/action

    Skilled Human SethBling

    77.21/ 100

    –/run · –/action

    Collaboration

    GPT-6.1 Terra

    54.86

    Skilled Human SethBling

    59.32

    Economy

    GPT-6.1 Terra

    34.66

    Skilled Human SethBling

    100.00

    Conflict

    GPT-6.1 Terra

    52.00

    Skilled Human SethBling

    90.00

    Logic

    GPT-6.1 Terra

    30.00

    Skilled Human SethBling

    30.00

    Creativity

    GPT-6.1 Terra

    93.40

    Skilled Human SethBling

    18.29

    OpenAIGPT-6.1 Terra

    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

    $56.62

    12,182 model calls · 68 failed calls

    Observed model

    openai/gpt-6.1-terra

    100% of observed calls

    PassedFailedJudged score

    Conflict

    Conflict · Zombie Apocalypse

    Can it stay alive under pressure?

    Result

    30 / 30

    Score / 100

    100.00

    Assessment

    strong
    Runs
    30
    LLM cost
    $2.59

    Conflict · Nightfall

    Can it survive a hard night in the open?

    Result

    20 / 30

    Score / 100

    66.67

    Assessment

    mixed
    By tier

    Locked night 9/15

    Night hunt 11/15

    Runs
    30
    LLM cost
    $8.45

    Conflict · Skywars vs Gemini 3.8 Flash

    Can it win fights against a fast, lighter opponent?

    Result

    19 / 30

    Score / 100

    63.33

    Assessment

    mixed
    Runs
    30
    LLM cost
    $2.59

    Conflict · Skywars vs Opus 5.5

    Can it win fights against a frontier opponent?

    Result

    11 / 30

    Score / 100

    36.67

    Assessment

    mixed
    Runs
    30
    LLM cost
    $2.18

    Conflict · Skywars vs GPT-6.1 Sol

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

    Result

    7 / 30

    Score / 100

    23.33

    Assessment

    weak
    Runs
    30
    LLM cost
    $1.94

    Economy

    Economy vs Gemini 3.8 Flash

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

    Result

    22 / 30

    Score / 100

    73.33

    Assessment

    strong
    Runs
    30
    LLM cost
    $7.15

    Economy vs Opus 5.5

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

    Result

    27 / 30

    Score / 100

    90.00

    Assessment

    strong
    Runs
    30
    LLM cost
    $7.46

    Economy vs GPT-6.1 Sol

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

    Result

    1 / 30

    Score / 100

    3.33

    Assessment

    weak
    Runs
    30
    LLM cost
    $7.93

    Logic

    Logic · Gate ladder

    Can it build redstone circuits that pass an exam?

    Result

    30 / 30

    Score / 100

    100.00

    Assessment

    strong
    By tier

    AND gate 15/15

    XOR gate 15/15

    Runs
    30
    LLM cost
    $3.42

    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
    $7.92

    Creativity

    Creativity

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

    Result

    25 / 25

    Score / 100

    93.40

    Assessment

    strong
    By build

    Godzilla 91.02

    Eiffel Tower 84.61

    Golden Gate Bridge 100.00

    Statue of Liberty 91.39

    Pelican riding a bicycle 100.00

    Runs
    25
    LLM cost
    $2.15

    Collaboration

    Collaboration · Ascent

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

    Result

    avg 30.20

    Score / 100

    50.33

    Assessment

    mixed
    Runs
    5
    LLM cost
    $0.39

    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 64.52

    Score / 100

    64.52

    Assessment

    strong
    Runs
    5
    LLM cost
    $0.55

    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 56.98

    Score / 100

    56.98

    Assessment

    mixed
    Runs
    5
    LLM cost
    $0.75

    Collaboration · Scavenger Hunt

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

    Result

    avg 51.40

    Score / 100

    47.59

    Assessment

    mixed
    Runs
    5
    LLM cost
    $1.16

    KradleBench

    Get notified when a new model is added