priorsBuilder

Meta · released 2026-08-13

Muse Spark 1.2

Measured on 70 benchmarks across 3 reasoning efforts.

General frontier

56.0

#25 of 136 models · best at max

includes estimates: 15% of the mix is measured for this model

Reasoning efforts in your mix

max56.0#58 of 313
minimal52.4#71 of 313
xhigh48.0#91 of 313

Your mix’s benchmarks

5 of 8 measured

reasoning

Humanity's Last Exam

587 results
  1. #48xhigh of 58745.5%

Leader: Opus 5.5 · 61.4%

knowledge

AA-Omniscience

526 results
  1. #42xhigh of 52627.2

Leader: Opus 5.5 · 46.4

reasoning

BullshitBench v2

196 results
  1. #73minimal of 19650.0%
  2. #75xhigh of 19649.0%

Leader: Opus 4.8 · 95.0%

coding

DeepSWE v1.1

70 results
  1. #39xhigh of 7054.9%

Leader: GPT-6 Astra · 74.1%

reasoning

SimpleBench

97 results
  1. #16max of 9774.5%

Leader: Opus 5.5 · 88.4%

All 70 benchmarks

show 65 more

agentic

AA-Briefcase v1.1

224 results
  1. #67xhigh of 2241328

Leader: Sonnet 5.5 · 1823

agentic

Agent Arena

50 results
  1. #38xhigh of 50-3.7%

Leader: Opus 5.5 · 15.5%

agentic

AutomationBench-AA

219 results
  1. #92xhigh of 21940.6%

Leader: Gemini 4 · 77.5%

agentic

BioMysteryBench

27 results
  1. #21xhigh of 2764.8%

Leader: Sonnet 5.5 · 81.1%

agentic

EnterpriseOps-Gym-AA

50 results
  1. #9xhigh of 5047.3%

Leader: Fable 5 · 51.1%

agentic

Harvey's Legal Agent Benchmark

79 results
  1. #1xhigh of 7925.4%

Leader: this model · 25.4%

agentic

Legal Research Bench

78 results
  1. #20xhigh of 7843.8%

Leader: Muse Spark 1.3 · 55.3%

agentic

SkillsBench

34 results
  1. #16xhigh of 3453.0%

Leader: DeepSeek V4.1 Flash · 69.8%

agentic

Tax Agent Bench

69 results
  1. #15xhigh of 6932.8%

Leader: Fable 5.1 · 49.2%

agentic

Terminal-Bench 4.0 (Vals)

47 results
  1. #40xhigh of 476.1%

Leader: Opus 5.5 · 65.2%

agentic

τ³-Banking

206 results
  1. #48xhigh of 20634.8%

Leader: Qwen 3.8 Max · 51.3%

coding

Arena WebDev

122 results
  1. #45xhigh of 1221533

Leader: Opus 5.5 · 1813

coding

Code Migration

77 results
  1. #42xhigh of 7730.0%

Leader: Sonnet 5.5 · 69.8%

coding

FrontierSWE v2

21 results
  1. #20max of 2112.0%

Leader: GPT-6 Astra · 65.5%

coding

IOI

44 results
  1. #41xhigh of 4421.8%

Leader: Gemini 4 · 100.0%

coding

ProgramBench v1 Almost Resolved

62 results
  1. #41xhigh of 622.5%

Leader: Opus 5.5 · 65.0%

coding

ProgramBench v1 Fully Resolved

62 results
  1. #20xhigh of 620.5%

Leader: Opus 5.5 · 18.5%

coding

ProgramBench v1 Raw Pass Rate

62 results
  1. #34xhigh of 6261.9%

Leader: Opus 5.5 · 87.0%

coding

SciCode

226 results
  1. #27xhigh of 22657.4%

Leader: Opus 5.5 · 66.9%

coding

SWE-bench Verified (Vals)

85 results
  1. #14xhigh of 8586.6%

Leader: Opus 5 · 97.0%

coding

Terminal-Bench 2.1 (Vals)

74 results
  1. #28xhigh of 7469.7%

Leader: Opus 5.5 · 87.6%

coding

Terminal-Bench v2.1-AA

238 results
  1. #44xhigh of 23880.1%

Leader: Fable 5.1 · 91.4%

coding

Terminal-Bench v4-AA

216 results
  1. #91xhigh of 2167.1%

Leader: Sonnet 5.5 · 63.6%

coding

Vibe Code Bench v1.1

105 results
  1. #25xhigh of 10579.1%

Leader: Sonnet 5.5 · 92.4%

coding

WeirdML v2

127 results
  1. #47xhigh of 12760.3%

Leader: GPT-6 Astra Pro · 93.6%

composite

AA Intelligence Index v4.3.2

610 results
  1. #65xhigh of 61039.6

Leader: Opus 5.5 · 57.6

composite

Arena Text

220 results
  1. #11xhigh of 2201492

Leader: Gemini 4 · 1525

composite

Arena Text — Multi-Turn

220 results
  1. #7xhigh of 2201506

Leader: Gemini 4 · 1553

composite

Vals Index

47 results
  1. #33xhigh of 4749.3%

Leader: Gemini 4 · 68.9%

composite

Vals Multimodal Index v1.2

33 results
  1. #6max of 3369.8%

Leader: Fable 5 · 74.2%

cost

AA Intelligence Index Cost per Task

177 results
  1. #112xhigh of 177$0.97

Leader: GPT-6 Luna · $0.00

cost

Cost to Run AA Intelligence Index

177 results
  1. #114xhigh of 177$1,385

Leader: GPT-6 Luna · $10.63

creative

MineBench

72 results
  1. #31max of 721604

Leader: GPT-6 Astra Pro · 2281

economics

CorpFin (Vals)

117 results
  1. #5xhigh of 11770.9%

Leader: Opus 5 · 73.2%

economics

Excel Modeling Benchmark

74 results
  1. #42xhigh of 7457.0%

Leader: Fable 5.1 · 76.7%

economics

Finance Agent v2

78 results
  1. #3xhigh of 7860.6%

Leader: Gemini 4 · 65.4%

economics

GDP.pdf

210 results
  1. #78xhigh of 21017.4%

Leader: Muse Wonder Full (Max) · 32.6%

economics

GDPval-AA v2.1

288 results
  1. #53xhigh of 2881474

Leader: Opus 5.5 · 1866

economics

MortgageTax (Vals)

84 results
  1. #40xhigh of 8465.4%

Leader: Opus 5 · 72.1%

economics

TaxEval (Vals)

125 results
  1. #1xhigh of 12580.4%

Leader: this model · 80.4%

knowledge

AA-Omniscience Accuracy

526 results
  1. #88xhigh of 52645.4%

Leader: Fable 5.1 · 67.2%

knowledge

AA-Omniscience Non-hallucination

526 results
  1. #56xhigh of 52666.7%

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

knowledge

GPQA Diamond

539 results
  1. #61xhigh of 53990.4%

Leader: GPT-6 Astra · 96.3%

knowledge

MedCode

94 results
  1. #23xhigh of 9449.3%

Leader: Opus 5 · 63.6%

knowledge

MedScribe

96 results
  1. #5xhigh of 9690.1%

Leader: Opus 5.5 · 91.4%

knowledge

MMLU Pro (Vals)

122 results
  1. #20xhigh of 12288.3%

Leader: Fable 5.1 · 92.4%

knowledge

MMMU Pro (Vals)

81 results
  1. #22xhigh of 8186.1%

Leader: Fable 5.1 · 90.6%

knowledge

Public Benefits Bench v1.1

50 results
  1. #11xhigh of 5068.5%

Leader: Opus 5 · 76.9%

long-context

AA-LCR v1.1

523 results
  1. #103xhigh of 52379.0%

Leader: Kimi K3 · 88.7%

long-context

MLCR-AA

94 results
  1. #20xhigh of 9431.1%

Leader: Sonnet 5.5 · 75.0%

math

ProofBench v1.1

52 results
  1. #35xhigh of 5243.0%

Leader: Sonnet 5.5 · 100.0%

other

Arena Vision

116 results
  1. #9xhigh of 1161293

Leader: Fable 5 · 1308

reasoning

CritPt

529 results
  1. #74xhigh of 52917.7%

Leader: Muse Wonder Full (Max) · 32.9%

reasoning

Humanity's Sixth Sense

25 results
  1. #21xhigh of 2524.5%

Leader: GPT-6 Astra · 53.6%

reasoning

LegalBench

135 results
  1. #22xhigh of 13585.3%

Leader: Fable 5 · 88.6%

reasoning

Roboflow Visual Reasoning

102 results
  1. #21high of 10275.7%
  2. #22low of 10275.1%

Leader: GPT-6 Astra · 91.2%

reasoning

SAGE

80 results
  1. #30xhigh of 8047.7%

Leader: Opus 4.7 · 56.1%

reliability

Tool-call error rate

171 results
  1. #53max of 1710.7%

Leader: GPT-5.4 Pro · 0.0%

reliability

Uptime (7d)

181 results
  1. #17max of 181100.0%

Leader: GPT-5 Pro · 100.0%

safety

CyberBench v1.1

45 results
  1. #23xhigh of 4569.5%

Leader: Wonder · 80.4%

safety

SRE Bench

32 results
  1. #30xhigh of 320.0%

Leader: GPT-6 Astra · 56.9%

speed

Latency (Time to First Token)

351 results
  1. #265xhigh of 35114s

Leader: Gemini 2.5 Flash-Lite · 0.3s

speed

Output Speed (tokens/s)

351 results
  1. #16xhigh of 351290.7

Leader: Celeris-1 · 1522.5

speed

Provider latency

177 results
  1. #160max of 1773.9s

Leader: Qwen 2.5 7B · 0.2s

speed

Provider throughput

176 results
  1. #25max of 176106.0

Leader: Muse Spark 1.1 · 204.3