priorsBuilder

Google · released 2026-07-21

Gemini 3.6 Flash

Measured on 80 benchmarks across 2 reasoning efforts.

General frontier

39.4

#29 of 32 models · best at high

Reasoning efforts in your mix

high39.4#88 of 129
minimal12.1#129 of 129

Your mix’s benchmarks

5 of 8 measured

reasoning

Humanity's Last Exam

587 results
  1. #78high of 58740.8%

Leader: Opus 5.5 · 61.4%

knowledge

AA-Omniscience

526 results
  1. #57high of 52622.1

Leader: Opus 5.5 · 46.4

reasoning

BullshitBench v2

196 results
  1. #101minimal of 19639.0%
  2. #130xhigh of 19628.0%

Leader: Opus 4.8 · 95.0%

coding

DeepSWE v1.1

70 results
  1. #53high of 7046.7%

Leader: GPT-6 Astra · 74.1%

math

FrontierMath Tier 4 (v2)

70 results
  1. #47high of 7022.0%

Leader: GPT-6.1 Sol · 100.0%

All 80 benchmarks

show 75 more

agentic

AA-Briefcase v1.1

224 results
  1. #124high of 224954

Leader: Sonnet 5.5 · 1823

agentic

Agent Arena

50 results
  1. #46high of 50-7.8%

Leader: Opus 5.5 · 15.5%

agentic

AutomationBench (Zapier)

115 results
  1. #44high of 11517.1%
  2. #48medium of 11516.4%
  3. #52low of 11515.7%

Leader: Gemini 4 · 51.3%

agentic

AutomationBench-AA

219 results
  1. #67high of 21953.0%

Leader: Gemini 4 · 77.5%

agentic

BioMysteryBench

27 results
  1. #27high of 2758.5%

Leader: Sonnet 5.5 · 81.1%

agentic

Harvey's Legal Agent Benchmark

79 results
  1. #42high of 793.3%

Leader: Muse Spark 1.2 · 25.4%

agentic

Legal Research Bench

78 results
  1. #53high of 7825.0%

Leader: Muse Spark 1.3 · 55.3%

agentic

Tax Agent Bench

69 results
  1. #47high of 6918.0%

Leader: Fable 5.1 · 49.2%

agentic

Terminal-Bench Science (Vals)

40 results
  1. #27high of 404.3%

Leader: GPT-6 Astra · 62.9%

agentic

τ³-Banking

206 results
  1. #66high of 20629.9%

Leader: Qwen 3.8 Max · 51.3%

coding

Arena Image-to-WebDev

58 results
  1. #27high of 581545

Leader: Opus 5.5 · 1749

coding

Arena WebDev

122 results
  1. #42high of 1221538

Leader: Opus 5.5 · 1813

coding

Code Migration

77 results
  1. #40high of 7730.9%

Leader: Sonnet 5.5 · 69.8%

coding

FrontierCode 1.1 Extended

132 results
  1. #96medium of 13248.0%
  2. #98high of 13247.6%
  3. #121low of 13232.8%

Leader: Opus 5.5 · 65.3%

coding

FrontierCode 1.1 Main

132 results
  1. #96medium of 13234.4%
  2. #98high of 13233.7%
  3. #118low of 13222.8%

Leader: Opus 5.5 · 54.6%

coding

IOI

44 results
  1. #39high of 4435.1%

Leader: Gemini 4 · 100.0%

coding

LiveCodeBench (Vals)

127 results
  1. #8high of 12788.1%

Leader: Fable 5.1 · 90.5%

coding

ProgramBench v1 Almost Resolved

62 results
  1. #39high of 622.5%

Leader: Opus 5.5 · 65.0%

coding

ProgramBench v1 Fully Resolved

62 results
  1. #45high of 620.0%

Leader: Opus 5.5 · 18.5%

coding

ProgramBench v1 Raw Pass Rate

62 results
  1. #36high of 6259.9%

Leader: Opus 5.5 · 87.0%

coding

SciCode

226 results
  1. #74high of 22653.4%

Leader: Opus 5.5 · 66.9%

coding

SWE-bench Verified (Vals)

85 results
  1. #25high of 8579.6%

Leader: Opus 5 · 97.0%

coding

Terminal-Bench 2.1 (Vals)

74 results
  1. #21high of 7473.8%

Leader: Opus 5.5 · 87.6%

coding

Terminal-Bench v2.1-AA

238 results
  1. #57high of 23877.5%

Leader: Fable 5.1 · 91.4%

coding

Terminal-Bench v4-AA

216 results
  1. #93high of 2167.1%

Leader: Sonnet 5.5 · 63.6%

coding

Vibe Code Bench v1.1

105 results
  1. #44high of 10564.0%

Leader: Sonnet 5.5 · 92.4%

coding

WeirdML v2

127 results
  1. #51xhigh of 12756.1%

Leader: GPT-6 Astra Pro · 93.6%

composite

AA Intelligence Index v4.3.2

610 results
  1. #94high of 61034.0

Leader: Opus 5.5 · 57.6

composite

Arena Text

220 results
  1. #22high of 2201482

Leader: Gemini 4 · 1525

composite

Arena Text — Multi-Turn

220 results
  1. #22high of 2201487

Leader: Gemini 4 · 1553

composite

Vals Multimodal Index v1.2

33 results
  1. #13high of 3365.1%

Leader: Fable 5 · 74.2%

cost

AA Intelligence Index Cost per Task

177 results
  1. #138high of 177$1.60

Leader: GPT-6 Luna · $0.00

cost

Cost to Run AA Intelligence Index

177 results
  1. #125high of 177$1,614

Leader: GPT-6 Luna · $10.63

creative

MineBench

72 results
  1. #28xhigh of 721656

Leader: GPT-6 Astra Pro · 2281

creative

threejseval

34 results
  1. #25high of 341356

Leader: Opus 5.5 · 2079

creative

VoxelBench (text)

55 results
  1. #18xhigh of 551788

Leader: GPT-6 Astra · 2733

economics

CorpFin (Vals)

117 results
  1. #46high of 11763.3%

Leader: Opus 5 · 73.2%

economics

Excel Modeling Benchmark

74 results
  1. #24high of 7465.4%

Leader: Fable 5.1 · 76.7%

economics

Finance Agent v2

78 results
  1. #18high of 7856.3%

Leader: Gemini 4 · 65.4%

economics

GDP.pdf

210 results
  1. #79high of 21017.4%

Leader: Muse Wonder Full (Max) · 32.6%

economics

GDPval-AA v2.1

288 results
  1. #100high of 2881283

Leader: Opus 5.5 · 1866

economics

MortgageTax (Vals)

84 results
  1. #23high of 8467.6%

Leader: Opus 5 · 72.1%

economics

TaxEval (Vals)

125 results
  1. #26high of 12574.9%

Leader: Muse Spark 1.2 · 80.4%

knowledge

AA-Omniscience Accuracy

526 results
  1. #60high of 52650.0%

Leader: Fable 5.1 · 67.2%

knowledge

AA-Omniscience Non-hallucination

526 results
  1. #140high of 52644.4%

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

knowledge

GPQA Diamond

539 results
  1. #31high of 53992.8%

Leader: GPT-6 Astra · 96.3%

knowledge

GPQA Diamond (Vals)

122 results
  1. #7high of 12293.4%

Leader: Gemini 3.1 Pro · 95.5%

knowledge

MedCode

94 results
  1. #11high of 9453.2%

Leader: Opus 5 · 63.6%

knowledge

MedScribe

96 results
  1. #56high of 9679.7%

Leader: Opus 5.5 · 91.4%

knowledge

MMLU Pro (Vals)

122 results
  1. #13high of 12289.3%

Leader: Fable 5.1 · 92.4%

knowledge

MMMU Pro (Vals)

81 results
  1. #7high of 8188.4%

Leader: Fable 5.1 · 90.6%

knowledge

MMMU-Pro

265 results
  1. #26high of 26583.2%

Leader: Opus 5.5 · 87.7%

knowledge

Public Benefits Bench v1.1

50 results
  1. #41high of 5056.6%

Leader: Opus 5 · 76.9%

long-context

AA-LCR v1.1

523 results
  1. #81high of 52380.0%

Leader: Kimi K3 · 88.7%

long-context

Arena Document

43 results
  1. #24high of 431456

Leader: Opus 5 · 1516

long-context

MLCR-AA

94 results
  1. #53high of 9414.4%

Leader: Sonnet 5.5 · 75.0%

math

FrontierMath Tiers 1–3 (v2)

115 results
  1. #38high of 11558.9%

Leader: GPT-6.1 Sol · 93.7%

other

Arena Vision

116 results
  1. #25high of 1161281

Leader: Fable 5 · 1308

other

T3 Code usage share

85 results
  1. #37xhigh of 850.0%

Leader: Opus 5.5 · 29.1%

reasoning

ARC-AGI-1

215 results
  1. #58high of 21591.2%
  2. #103medium of 21583.2%
  3. #113low of 21576.5%
  4. #172minimal of 21534.5%

Leader: Fable 5 · 98.5%

reasoning

ARC-AGI-2

216 results
  1. #78high of 21660.4%
  2. #96medium of 21650.4%
  3. #117low of 21630.4%
  4. #169minimal of 2162.6%

Leader: GPT-6 Astra · 95.0%

reasoning

Blueprint-Bench 2

28 results
  1. #15xhigh of 2831.2%

Leader: Gemini 4 · 54.4%

reasoning

CritPt

529 results
  1. #112high of 52910.6%

Leader: Muse Wonder Full (Max) · 32.9%

reasoning

Furniture Assembly

31 results
  1. #27high of 3123.3%

Leader: Opus 5.5 · 83.3%

reasoning

Humanity's Sixth Sense

25 results
  1. #11high of 2531.9%

Leader: GPT-6 Astra · 53.6%

reasoning

LegalBench

135 results
  1. #11high of 13586.7%

Leader: Fable 5 · 88.6%

reasoning

Roboflow Visual Reasoning

102 results
  1. #13high of 10281.0%
  2. #17low of 10277.7%

Leader: GPT-6 Astra · 91.2%

reasoning

SAGE

80 results
  1. #25high of 8048.7%

Leader: Opus 4.7 · 56.1%

reliability

Tool-call error rate

172 results
  1. #161xhigh of 1728.1%

Leader: Gemini 4 · 0.0%

reliability

Uptime (7d)

182 results
  1. #109xhigh of 18299.0%

Leader: GPT-5 Pro · 100.0%

safety

CyberBench v1.1

45 results
  1. #39high of 4545.4%

Leader: Wonder · 80.4%

speed

Latency (Time to First Token)

351 results
  1. #267high of 35115s

Leader: Gemini 2.5 Flash Lite · 0.3s

speed

Output Speed (tokens/s)

351 results
  1. #42high of 351201.1

Leader: Celeris-1 · 1564.1

speed

Provider latency

178 results
  1. #106xhigh of 1781.5s

Leader: Qwen 2.5 7B · 0.2s

speed

Provider throughput

176 results
  1. #7xhigh of 176132.0

Leader: Muse Spark 1.1 · 204.5