priorsBuilder

xAI · released 2026-04-17

Grok 4.3

Measured on 74 benchmarks across 6 reasoning efforts.

General frontier

50.7

#34 of 136 models · best at minimal

includes estimates: 5% of the mix is measured for this model

Reasoning efforts in your mix

minimal50.7#74 of 313
xhigh46.5#98 of 313
high36.9#169 of 313
medium34.2#189 of 313
low28.5#233 of 313
none16.0#291 of 313

Your mix’s benchmarks

4 of 8 measured

reasoning

Humanity's Last Exam

587 results
  1. #103high of 58737.2%
  2. #148medium of 58730.0%
  3. #224low of 58718.4%
  4. #368none of 5876.8%

Leader: Opus 5.5 · 61.4%

knowledge

AA-Omniscience

526 results
  1. #74high of 52618.0
  2. #75medium of 52616.7
  3. #84low of 52613.9
  4. #282none of 526-33.1

Leader: Opus 5.5 · 46.4

reasoning

BullshitBench v2

196 results
  1. #71minimal of 19650.0%
  2. #85xhigh of 19646.0%

Leader: Opus 4.8 · 95.0%

math

FrontierMath Tier 4 (v2)

70 results
  1. #55high of 7014.6%

Leader: GPT-6.1 Sol · 100.0%

All 74 benchmarks

show 70 more

agentic

AA-AnalystAgent

44 results
  1. #40high of 448.8%

Leader: Gemini 3.7 Flash · 60.0%

agentic

AA-Briefcase v1.1

224 results
  1. #152high of 224762
  2. #158none of 224724

Leader: Sonnet 5.5 · 1823

agentic

APEX-Agents-AA

33 results
  1. #21high of 3317.0%

Leader: Gemini 3.5 Flash · 47.1%

agentic

AutomationBench-AA

219 results
  1. #185high of 2191.4%
  2. #193none of 2190.9%

Leader: Gemini 4 · 77.5%

agentic

EnterpriseOps-Gym-AA

50 results
  1. #47high of 5026.3%

Leader: Fable 5 · 51.1%

agentic

Harvey's Legal Agent Benchmark

79 results
  1. #60high of 790.4%

Leader: Muse Spark 1.2 · 25.4%

agentic

ITBench-AA

45 results
  1. #30high of 4532.7%

Leader: GPT-5.6 Sol · 56.2%

agentic

Legal Research Bench

78 results
  1. #61high of 7815.4%

Leader: Muse Spark 1.3 · 55.3%

agentic

SkillsBench

34 results
  1. #28high of 3440.6%

Leader: DeepSeek V4.1 Flash · 69.8%

agentic

Tax Agent Bench

69 results
  1. #63high of 696.2%

Leader: Fable 5.1 · 49.2%

agentic

Vending-Bench 2

67 results
  1. #62xhigh of 67$35.26

Leader: GPT-6 Astra · $15,515

agentic

τ²-Bench Telecom

393 results
  1. #10high of 39397.7%
  2. #59medium of 39391.2%
  3. #72low of 39388.9%
  4. #171none of 39365.8%

Leader: GLM 5.2 · 99.1%

agentic

τ³-Banking

206 results
  1. #133high of 20612.4%
  2. #159none of 2068.0%

Leader: Qwen 3.8 Max · 51.3%

coding

Arena Image-to-WebDev

58 results
  1. #55xhigh of 581375

Leader: Opus 5.5 · 1749

coding

Arena WebDev

122 results
  1. #98xhigh of 1221357

Leader: Opus 5.5 · 1813

coding

Code Migration

77 results
  1. #67high of 776.8%

Leader: Sonnet 5.5 · 69.8%

coding

LiveCodeBench (Vals)

127 results
  1. #39high of 12784.5%

Leader: Fable 5.1 · 90.5%

coding

ProgramBench v1 Almost Resolved

62 results
  1. #58high of 620.0%

Leader: Opus 5.5 · 65.0%

coding

ProgramBench v1 Fully Resolved

62 results
  1. #47high of 620.0%

Leader: Opus 5.5 · 18.5%

coding

ProgramBench v1 Raw Pass Rate

62 results
  1. #55high of 6222.1%

Leader: Opus 5.5 · 87.0%

coding

SciCode

226 results
  1. #120high of 22648.3%
  2. #174none of 22639.4%

Leader: Opus 5.5 · 66.9%

coding

SWE-bench Verified (Vals)

85 results
  1. #54high of 8571.4%

Leader: Opus 5 · 97.0%

coding

Terminal-Bench 2.0 (Vals)

63 results
  1. #27xhigh of 6343.4%

Leader: GPT-5.5 · 73.2%

coding

Terminal-Bench 2.1 (Vals)

74 results
  1. #65high of 7441.9%

Leader: Opus 5.5 · 87.6%

coding

Terminal-Bench Hard-AA

387 results
  1. #64high of 38737.9%
  2. #114medium of 38730.3%
  3. #130low of 38726.5%
  4. #169none of 38718.9%

Leader: GPT-5.6 Sol · 65.9%

coding

Terminal-Bench v2.1-AA

238 results
  1. #140high of 23839.7%
  2. #151none of 23834.1%

Leader: Fable 5.1 · 91.4%

coding

Terminal-Bench v4-AA

216 results
  1. #154high of 2160.0%
  2. #176none of 2160.0%

Leader: Sonnet 5.5 · 63.6%

coding

Vibe Code Bench v1.1

105 results
  1. #76high of 10519.4%

Leader: Sonnet 5.5 · 92.4%

coding

WeirdML v2

127 results
  1. #62xhigh of 12749.9%

Leader: GPT-6 Astra Pro · 93.6%

composite

AA Intelligence Index v4.3.2

610 results
  1. #164high of 61024.9
  2. #165medium of 61024.8
  3. #172low of 61024.3
  4. #297none of 61014.0

Leader: Opus 5.5 · 57.6

composite

Arena Text

220 results
  1. #78xhigh of 2201442

Leader: Gemini 4 · 1525

composite

Arena Text — Multi-Turn

220 results
  1. #83xhigh of 2201446

Leader: Gemini 4 · 1553

composite

Vals Multimodal Index v1.2

33 results
  1. #29high of 3343.3%

Leader: Fable 5 · 74.2%

cost

AA Intelligence Index Cost per Task

177 results
  1. #52none of 177$0.17
  2. #57high of 177$0.21

Leader: GPT-6 Luna · $0.00

cost

Cost to Run AA Intelligence Index

177 results
  1. #46none of 177$250
  2. #59high of 177$363

Leader: GPT-6 Luna · $10.63

creative

MineBench

72 results
  1. #55xhigh of 721196

Leader: GPT-6 Astra Pro · 2281

economics

CorpFin (Vals)

117 results
  1. #8xhigh of 11768.5%

Leader: Opus 5 · 73.2%

economics

Excel Modeling Benchmark

74 results
  1. #69high of 7418.4%

Leader: Fable 5.1 · 76.7%

economics

Finance Agent v2

78 results
  1. #64high of 7837.7%

Leader: Gemini 4 · 65.4%

economics

GDP.pdf

210 results
  1. #152high of 2105.8%
  2. #178none of 2102.8%

Leader: Muse Wonder Full (Max) · 32.6%

economics

GDPval-AA v2.1

288 results
  1. #171none of 288942
  2. #172high of 288937

Leader: Opus 5.5 · 1866

economics

MortgageTax (Vals)

84 results
  1. #75xhigh of 8448.3%

Leader: Opus 5 · 72.1%

economics

Public Benefits Bench v1 (Vals)

13 results
  1. #11high of 1350.1%

Leader: Fable 5 · 71.7%

economics

TaxEval (Vals)

125 results
  1. #81xhigh of 12570.8%

Leader: Muse Spark 1.2 · 80.4%

instruction

IFBench

399 results
  1. #1medium of 39983.3%
  2. #5high of 39981.3%
  3. #6low of 39981.0%
  4. #185none of 39947.6%

Leader: this model · 83.3%

knowledge

AA-Omniscience Accuracy

526 results
  1. #155high of 52634.8%
  2. #190medium of 52628.8%
  3. #207low of 52626.7%
  4. #242none of 52623.7%

Leader: Fable 5.1 · 67.2%

knowledge

AA-Omniscience Non-hallucination

526 results
  1. #7medium of 52683.1%
  2. #9low of 52682.6%
  3. #26high of 52674.2%
  4. #236none of 52625.5%

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

knowledge

Arena Search

28 results
  1. #20xhigh of 281165

Leader: GPT-5.6 Sol · 1257

knowledge

CaseLaw (Vals)

53 results
  1. #1xhigh of 5379.3%

Leader: this model · 79.3%

knowledge

GPQA Diamond

539 results
  1. #64high of 53990.1%
  2. #83medium of 53989.0%
  3. #145low of 53984.3%
  4. #327none of 53965.8%

Leader: GPT-6 Astra · 96.3%

knowledge

GPQA Diamond (Vals)

122 results
  1. #22xhigh of 12291.4%

Leader: Gemini 3.1 Pro · 95.5%

knowledge

MedCode

94 results
  1. #66xhigh of 9438.1%

Leader: Opus 5 · 63.6%

knowledge

MedScribe

96 results
  1. #73xhigh of 9674.4%

Leader: Opus 5.5 · 91.4%

knowledge

MMLU Pro (Vals)

122 results
  1. #52high of 12285.8%

Leader: Fable 5.1 · 92.4%

knowledge

MMMU Pro (Vals)

81 results
  1. #34xhigh of 8183.1%

Leader: Fable 5.1 · 90.6%

knowledge

MMMU-Pro

265 results
  1. #73high of 26578.1%
  2. #92medium of 26575.8%
  3. #125low of 26572.8%
  4. #175none of 26564.8%

Leader: Opus 5.5 · 87.7%

knowledge

Public Benefits Bench v1.1

50 results
  1. #45high of 5051.7%

Leader: Opus 5 · 76.9%

long-context

AA-LCR v1.1

523 results
  1. #155medium of 52375.0%
  2. #168low of 52374.0%
  3. #182high of 52373.0%
  4. #366none of 52332.3%

Leader: Kimi K3 · 88.7%

long-context

MLCR-AA

94 results
  1. #65high of 9410.0%

Leader: Sonnet 5.5 · 75.0%

math

FrontierMath Tiers 1–3 (v2)

115 results
  1. #55high of 11542.8%

Leader: GPT-6.1 Sol · 93.7%

other

Arena Vision

116 results
  1. #59xhigh of 1161243

Leader: Fable 5 · 1308

reasoning

CritPt

529 results
  1. #133high of 5298.0%
  2. #155medium of 5294.9%
  3. #247low of 5290.6%
  4. #337none of 5290.0%

Leader: Muse Wonder Full (Max) · 32.9%

reasoning

LegalBench

135 results
  1. #31xhigh of 13584.5%

Leader: Fable 5 · 88.6%

reasoning

SAGE

80 results
  1. #80xhigh of 8019.7%

Leader: Opus 4.7 · 56.1%

reliability

Tool-call error rate

171 results
  1. #43xhigh of 1710.5%

Leader: GPT-5.4 Pro · 0.0%

reliability

Uptime (7d)

181 results
  1. #119xhigh of 18198.9%

Leader: GPT-5 Pro · 100.0%

speed

Latency (Time to First Token)

351 results
  1. #18none of 3510.7s
  2. #238low of 3516.7s
  3. #263medium of 35114s
  4. #291high of 35125s

Leader: Gemini 2.5 Flash-Lite · 0.3s

speed

Output Speed (tokens/s)

351 results
  1. #129none of 351115.9
  2. #147high of 351107.9
  3. #159medium of 351101.9
  4. #172low of 35195.3

Leader: Celeris-1 · 1522.5

speed

Provider latency

177 results
  1. #61xhigh of 1771.0s

Leader: Qwen 2.5 7B · 0.2s

speed

Provider throughput

176 results
  1. #40xhigh of 17690.0

Leader: Muse Spark 1.1 · 204.3