priorsBuilder

Moonshot · released 2026-07-16

Kimi K3

Measured on 99 benchmarks across 2 reasoning efforts.

General frontier

46.8

#18 of 32 models · best at max

Reasoning efforts in your mix

max46.8#59 of 128
low34.3#110 of 128

Your mix’s benchmarks

6 of 8 measured

reasoning

Humanity's Last Exam

587 results
  1. #41max of 58746.9%
  2. #180low of 58725.0%

Leader: Opus 5.5 · 61.4%

knowledge

AA-Omniscience

526 results
  1. #68max of 52619.7
  2. #110low of 5263.9

Leader: Opus 5.5 · 46.4

reasoning

BullshitBench v2

196 results
  1. #19xhigh of 19674.0%
  2. #23minimal of 19672.0%

Leader: Opus 4.8 · 95.0%

coding

DeepSWE v1.1

70 results
  1. #19max of 7068.5%

Leader: GPT-6 Astra · 74.1%

math

FrontierMath Tier 4 (v2)

70 results
  1. #28max of 7039.0%

Leader: GPT-6.1 Sol · 100.0%

reasoning

SimpleBench

97 results
  1. #35max of 9760.7%

Leader: Opus 5.5 · 88.4%

All 99 benchmarks

show 93 more

agentic

AA-AnalystAgent

44 results
  1. #20max of 4438.8%

Leader: Gemini 3.7 Flash · 60.0%

agentic

AA-Briefcase v1.1

224 results
  1. #34max of 2241501
  2. #121low of 224986

Leader: Sonnet 5.5 · 1823

agentic

Agent Arena

50 results
  1. #18max of 503.8%

Leader: Opus 5.5 · 15.5%

agentic

APEX-Agents-AA

33 results
  1. #3max of 3341.3%

Leader: Gemini 3.5 Flash · 47.1%

agentic

AutomationBench (Zapier)

115 results
  1. #32max of 11522.7%

Leader: Gemini 4 · 51.3%

agentic

AutomationBench-AA

219 results
  1. #41max of 21958.3%
  2. #77low of 21949.3%

Leader: Gemini 4 · 77.5%

agentic

BioMysteryBench

27 results
  1. #9max of 2772.6%

Leader: Sonnet 5.5 · 81.1%

agentic

EnterpriseOps-Gym-AA

50 results
  1. #13max of 5045.3%

Leader: Fable 5 · 51.1%

agentic

Harvey's Legal Agent Benchmark

79 results
  1. #12max of 7912.9%

Leader: Muse Spark 1.2 · 25.4%

agentic

ITBench-AA

45 results
  1. #9max of 4547.7%

Leader: GPT-5.6 Sol · 56.2%

agentic

Legal Research Bench

78 results
  1. #16max of 7846.2%

Leader: Muse Spark 1.3 · 55.3%

agentic

PostTrainBench v1.1

12 results
  1. #6max of 1232.0%

Leader: Fable 5 · 41.8%

agentic

Tax Agent Bench

69 results
  1. #13max of 6933.4%

Leader: Fable 5.1 · 49.2%

agentic

Terminal-Bench 4.0 (Vals)

47 results
  1. #29max of 4717.2%

Leader: Opus 5.5 · 65.2%

agentic

Terminal-Bench Science (Vals)

40 results
  1. #36max of 401.4%

Leader: GPT-6 Astra · 62.9%

agentic

Time Horizon Index: KSP

13 results
  1. #9max of 1310.5%

Leader: Opus 5.5 · 91.3%

agentic

Vending-Bench 2

67 results
  1. #28max of 67$5,165

Leader: GPT-6 Astra · $15,515

agentic

Vending-Bench Arena

12 results
  1. #12max of 12$3,200

Leader: GPT-6 Astra · $12,400

agentic

WeirdML v3

17 results
  1. #13high of 177.2%

Leader: GPT-6 Astra · 42.2%

agentic

τ³-Banking

206 results
  1. #11max of 20646.0%
  2. #25low of 20641.6%

Leader: Qwen 3.8 Max · 51.3%

coding

Arena Image-to-WebDev

58 results
  1. #20max of 581579

Leader: Opus 5.5 · 1749

coding

Arena WebDev

122 results
  1. #13max of 1221654

Leader: Opus 5.5 · 1813

coding

Code Migration

77 results
  1. #54max of 7716.1%

Leader: Sonnet 5.5 · 69.8%

coding

FrontierCode 1.1 Extended

130 results
  1. #44none of 13058.2%

Leader: Opus 5.5 · 65.3%

coding

FrontierCode 1.1 Main

130 results
  1. #49none of 13044.2%

Leader: Opus 5.5 · 54.6%

coding

FrontierSWE v2

21 results
  1. #12max of 2125.9%

Leader: GPT-6 Astra · 65.5%

coding

IOI

44 results
  1. #26max of 4448.9%

Leader: Gemini 4 · 100.0%

coding

LiveCodeBench (Vals)

127 results
  1. #18max of 12787.2%

Leader: Fable 5.1 · 90.5%

coding

ProgramBench v1 Almost Resolved

62 results
  1. #14max of 6217.5%

Leader: Opus 5.5 · 65.0%

coding

ProgramBench v1 Fully Resolved

62 results
  1. #10max of 622.0%

Leader: Opus 5.5 · 18.5%

coding

ProgramBench v1 Raw Pass Rate

62 results
  1. #32max of 6262.8%

Leader: Opus 5.5 · 87.0%

coding

SciCode

226 results
  1. #14max of 22659.5%
  2. #80low of 22652.7%

Leader: Opus 5.5 · 66.9%

coding

SWE-bench Verified (Vals)

85 results
  1. #8max of 8593.4%

Leader: Opus 5 · 97.0%

coding

Terminal-Bench 2.1 (Vals)

74 results
  1. #9max of 7480.9%

Leader: Opus 5.5 · 87.6%

coding

Terminal-Bench v2.1-AA

238 results
  1. #27max of 23885.0%
  2. #38low of 23882.4%

Leader: Fable 5.1 · 91.4%

coding

Terminal-Bench v4-AA

216 results
  1. #76max of 21612.6%
  2. #78low of 21612.6%

Leader: Sonnet 5.5 · 63.6%

coding

Vibe Code Bench 1-100

22 results
  1. #11max of 2218.2%

Leader: Opus 5.5 · 30.4%

coding

Vibe Code Bench v1.1

105 results
  1. #15max of 10585.0%

Leader: Sonnet 5.5 · 92.4%

coding

WeirdML v2

127 results
  1. #17max of 12782.6%

Leader: GPT-6 Astra Pro · 93.6%

composite

AA Intelligence Index v4.3.2

610 results
  1. #44max of 61043.6
  2. #119low of 61030.1

Leader: Opus 5.5 · 57.6

composite

Arena Text

220 results
  1. #15max of 2201488

Leader: Gemini 4 · 1525

composite

Arena Text — Multi-Turn

220 results
  1. #9max of 2201500

Leader: Gemini 4 · 1553

composite

Vals Index

47 results
  1. #30max of 4750.3%

Leader: Gemini 4 · 68.9%

composite

Vals Multimodal Index v1.2

33 results
  1. #3max of 3373.4%

Leader: Fable 5 · 74.2%

composite

Vals RSI Index v1.1

24 results
  1. #14max of 2420.9%

Leader: Opus 5.5 · 37.3%

composite

Vals RSI Index v1.1: Harness Engineering: Judge

24 results
  1. #17max of 249.9%

Leader: GPT-6 Sol · 22.8%

composite

Vals RSI Index v1.1: Post-training: Finance Agent

24 results
  1. #16max of 240.0%

Leader: Opus 5.5 · 35.1%

composite

Vals RSI Index v1.1: Pre-training: Compression

24 results
  1. #14max of 2433.3%

Leader: Opus 5.5 · 46.5%

composite

Vals RSI Index v1.1: Pre-training: LM Training

24 results
  1. #13max of 2440.4%

Leader: Opus 5.5 · 53.0%

cost

AA Intelligence Index Cost per Task

177 results
  1. #124low of 177$1.15
  2. #147max of 177$2.00

Leader: GPT-6 Luna · $0.00

cost

Cost to Run AA Intelligence Index

177 results
  1. #108low of 177$1,190
  2. #155max of 177$3,658

Leader: GPT-6 Luna · $10.63

creative

MineBench

72 results
  1. #22max of 721746

Leader: GPT-6 Astra Pro · 2281

creative

threejseval

34 results
  1. #15high of 341528
  2. #17low of 341473

Leader: Opus 5.5 · 2079

creative

VoxelBench (text)

55 results
  1. #11max of 551954

Leader: GPT-6 Astra · 2732

economics

CorpFin (Vals)

117 results
  1. #3max of 11771.6%

Leader: Opus 5 · 73.2%

economics

Excel Modeling Benchmark

74 results
  1. #20max of 7466.7%

Leader: Fable 5.1 · 76.7%

economics

Finance Agent v2

78 results
  1. #33max of 7853.1%

Leader: Gemini 4 · 65.4%

economics

GDP.pdf

210 results
  1. #50max of 21022.0%
  2. #72low of 21018.6%

Leader: Muse Wonder Full (Max) · 32.6%

economics

GDPval-AA v2.1

288 results
  1. #45max of 2881536
  2. #127low of 2881126

Leader: Opus 5.5 · 1866

economics

MortgageTax (Vals)

84 results
  1. #32max of 8466.3%

Leader: Opus 5 · 72.1%

economics

TaxEval (Vals)

125 results
  1. #12max of 12575.7%

Leader: Muse Spark 1.2 · 80.4%

knowledge

AA-Omniscience Accuracy

526 results
  1. #73max of 52647.6%
  2. #84low of 52645.8%

Leader: Fable 5.1 · 67.2%

knowledge

AA-Omniscience Non-hallucination

526 results
  1. #132max of 52646.8%
  2. #254low of 52622.9%

Leader: MiniCPM5-1B (Non-reasoning) · 99.1%

knowledge

GPQA Diamond

539 results
  1. #16max of 53993.5%
  2. #146low of 53984.2%

Leader: GPT-6 Astra · 96.3%

knowledge

GPQA Diamond (Vals)

122 results
  1. #12max of 12292.9%

Leader: Gemini 3.1 Pro · 95.5%

knowledge

Harvey LAB-AA v1.1

27 results
  1. #8max of 275.3%

Leader: Muse Wonder Full (Max) · 13.1%

knowledge

MedCode

94 results
  1. #22max of 9449.4%

Leader: Opus 5 · 63.6%

knowledge

MedScribe

96 results
  1. #14max of 9688.0%

Leader: Opus 5.5 · 91.4%

knowledge

MMLU Pro (Vals)

122 results
  1. #22max of 12288.0%

Leader: Fable 5.1 · 92.4%

knowledge

MMMU Pro (Vals)

81 results
  1. #11max of 8188.2%

Leader: Fable 5.1 · 90.6%

knowledge

MMMU-Pro

265 results
  1. #43max of 26580.5%
  2. #70low of 26578.5%

Leader: Opus 5.5 · 87.7%

knowledge

Public Benefits Bench v1.1

50 results
  1. #12max of 5068.2%

Leader: Opus 5 · 76.9%

long-context

AA-LCR v1.1

523 results
  1. #1max of 52388.7%
  2. #98low of 52379.3%

Leader: this model · 88.7%

long-context

MLCR-AA

94 results
  1. #16max of 9438.3%

Leader: Sonnet 5.5 · 75.0%

math

FrontierMath Tiers 1–3 (v2)

115 results
  1. #22max of 11572.2%

Leader: GPT-6.1 Sol · 93.7%

math

ProofBench v1.1

52 results
  1. #12max of 5287.0%

Leader: Sonnet 5.5 · 100.0%

reasoning

ARC-AGI-1

215 results
  1. #37max of 21594.5%
  2. #89high of 21586.7%
  3. #131low of 21565.7%

Leader: Fable 5 · 98.5%

reasoning

ARC-AGI-2

216 results
  1. #77max of 21660.4%
  2. #90high of 21655.0%
  3. #136low of 21612.4%

Leader: GPT-6 Astra · 95.0%

reasoning

Blueprint-Bench 2

28 results
  1. #19max of 2829.5%

Leader: Gemini 4 · 54.4%

reasoning

CritPt

529 results
  1. #49max of 52923.4%
  2. #171low of 5293.1%

Leader: Muse Wonder Full (Max) · 32.9%

reasoning

Furniture Assembly

31 results
  1. #20max of 3134.2%

Leader: Opus 5.5 · 83.3%

reasoning

Humanity's Sixth Sense

25 results
  1. #17max of 2525.5%

Leader: GPT-6 Astra · 53.6%

reasoning

LegalBench

135 results
  1. #15max of 13586.2%

Leader: Fable 5 · 88.6%

reasoning

Roboflow Visual Reasoning

102 results
  1. #28high of 10274.2%
  2. #88low of 10242.4%

Leader: GPT-6 Astra · 91.2%

reasoning

SAGE

80 results
  1. #5max of 8052.8%

Leader: Opus 4.7 · 56.1%

reliability

Tool-call error rate

171 results
  1. #34max of 1710.3%

Leader: GPT-5.4 Pro · 0.0%

reliability

Uptime (7d)

181 results
  1. #140max of 18198.3%

Leader: GPT-5 Pro · 100.0%

safety

CWE-Bench

19 results
  1. #11max of 1953.3%

Leader: Grok 4.7 · 68.3%

safety

CyberBench v1.1

45 results
  1. #8max of 4575.2%

Leader: Wonder · 80.4%

speed

Latency (Time to First Token)

351 results
  1. #207max of 3513.2s
  2. #211low of 3513.2s

Leader: Gemini 2.5 Flash-Lite · 0.3s

speed

Output Speed (tokens/s)

351 results
  1. #303low of 35146.2
  2. #310max of 35143.7

Leader: Celeris-1 · 1522.5

speed

Provider latency

177 results
  1. #98max of 1771.4s

Leader: Qwen 2.5 7B · 0.2s

speed

Provider throughput

176 results
  1. #132max of 17646.0

Leader: Muse Spark 1.1 · 204.3