priorsBuilder

Google · released 2026-05-19

Gemini 3.5 Flash

Measured on 91 benchmarks across 4 reasoning efforts.

General frontier

43.5

#24 of 32 models · best at xhigh

Reasoning efforts in your mix

xhigh43.5#76 of 129
medium38.9#91 of 129
high38.6#93 of 129
minimal28.9#119 of 129

Your mix’s benchmarks

6 of 8 measured

reasoning

Humanity's Last Exam

587 results
  1. #62high of 58742.7%
  2. #73medium of 58741.3%
  3. #182minimal of 58724.1%

Leader: Opus 5.5 · 61.4%

knowledge

AA-Omniscience

526 results
  1. #60high of 52621.2
  2. #63medium of 52620.8
  3. #127minimal of 5260.7

Leader: Opus 5.5 · 46.4

reasoning

BullshitBench v2

196 results
  1. #147xhigh of 19620.0%
  2. #151minimal of 19619.0%

Leader: Opus 4.8 · 95.0%

coding

DeepSWE v1.1

70 results
  1. #61high of 7036.1%

Leader: GPT-6 Astra · 74.1%

math

FrontierMath Tier 4 (v2)

70 results
  1. #42high of 7026.8%

Leader: GPT-6.1 Sol · 100.0%

reasoning

SimpleBench

97 results
  1. #12xhigh of 9776.7%

Leader: Opus 5.5 · 88.4%

All 91 benchmarks

show 85 more

agentic

AA-AnalystAgent

44 results
  1. #15high of 4445.0%

Leader: Gemini 3.7 Flash · 60.0%

agentic

AA-Briefcase v1.1

224 results
  1. #139medium of 224877
  2. #141high of 224874

Leader: Sonnet 5.5 · 1823

agentic

APEX-Agents-AA

33 results
  1. #1high of 3347.1%

Leader: this model · 47.1%

agentic

AutomationBench (Zapier)

115 results
  1. #59high of 11513.4%
  2. #69medium of 11511.4%
  3. #78low of 1159.0%
  4. #110minimal of 1151.2%

Leader: Gemini 4 · 51.3%

agentic

AutomationBench-AA

219 results
  1. #89high of 21942.1%

Leader: Gemini 4 · 77.5%

agentic

EnterpriseOps-Gym-AA

50 results
  1. #3high of 5050.1%

Leader: Fable 5 · 51.1%

agentic

Harvey's Legal Agent Benchmark

79 results
  1. #45high of 792.5%

Leader: Muse Spark 1.2 · 25.4%

agentic

ITBench-AA

45 results
  1. #17high of 4540.3%

Leader: GPT-5.6 Sol · 56.2%

agentic

Legal Research Bench

78 results
  1. #44high of 7830.8%

Leader: Muse Spark 1.3 · 55.3%

agentic

SkillsBench

34 results
  1. #17high of 3452.7%

Leader: DeepSeek V4.1 Flash · 69.8%

agentic

Tax Agent Bench

69 results
  1. #42high of 6921.7%

Leader: Fable 5.1 · 49.2%

agentic

Terminal-Bench 4.0 (Vals)

47 results
  1. #39high of 476.1%

Leader: Opus 5.5 · 65.2%

agentic

Terminal-Bench Science (Vals)

40 results
  1. #20high of 405.7%

Leader: GPT-6 Astra · 62.9%

agentic

Vending-Bench 2

67 results
  1. #27xhigh of 67$5,396

Leader: GPT-6 Astra · $15,515

agentic

Web Search Index

4 results
  1. #3high of 441.4%

Leader: Fable 5 · 48.5%

agentic

τ²-Bench Telecom

393 results
  1. #19medium of 39395.6%
  2. #23high of 39395.3%
  3. #186minimal of 39358.8%

Leader: GLM 5.2 · 99.1%

agentic

τ³-Banking

206 results
  1. #57high of 20632.2%

Leader: Qwen 3.8 Max · 51.3%

coding

Arena Image-to-WebDev

58 results
  1. #39xhigh of 581512
  2. #40medium of 581505
  3. #41high of 581501

Leader: Opus 5.5 · 1749

coding

Arena WebDev

122 results
  1. #51xhigh of 1221511
  2. #55high of 1221499
  3. #57medium of 1221491

Leader: Opus 5.5 · 1813

coding

Code Migration

77 results
  1. #45high of 7726.7%

Leader: Sonnet 5.5 · 69.8%

coding

DeepSWE v1

29 results
  1. #15medium of 2928.3%

Leader: GPT-5.5 · 70.0%

coding

LiveCodeBench (Vals)

127 results
  1. #12high of 12787.6%

Leader: Fable 5.1 · 90.5%

coding

ProgramBench v1 Almost Resolved

62 results
  1. #43high of 621.5%

Leader: Opus 5.5 · 65.0%

coding

ProgramBench v1 Fully Resolved

62 results
  1. #43high of 620.0%

Leader: Opus 5.5 · 18.5%

coding

ProgramBench v1 Raw Pass Rate

62 results
  1. #39high of 6256.4%

Leader: Opus 5.5 · 87.0%

coding

SciCode

226 results
  1. #70high of 22653.9%

Leader: Opus 5.5 · 66.9%

coding

SWE-bench Verified (Vals)

85 results
  1. #27high of 8578.8%

Leader: Opus 5 · 97.0%

coding

Terminal-Bench 2.0 (Vals)

63 results
  1. #4high of 6367.4%

Leader: GPT-5.5 · 73.2%

coding

Terminal-Bench 2.1 (Vals)

74 results
  1. #20high of 7474.2%

Leader: Opus 5.5 · 87.6%

coding

Terminal-Bench Hard-AA

387 results
  1. #33minimal of 38746.2%
  2. #53high of 38740.9%
  3. #58medium of 38739.4%

Leader: GPT-5.6 Sol · 65.9%

coding

Terminal-Bench v2.1-AA

238 results
  1. #51high of 23878.7%

Leader: Fable 5.1 · 91.4%

coding

Terminal-Bench v4-AA

216 results
  1. #94high of 2166.6%

Leader: Sonnet 5.5 · 63.6%

coding

Vibe Code Bench v1.1

105 results
  1. #52high of 10548.7%

Leader: Sonnet 5.5 · 92.4%

coding

WeirdML v2

127 results
  1. #40xhigh of 12762.6%

Leader: GPT-6 Astra Pro · 93.6%

composite

AA Intelligence Index v4.3.2

610 results
  1. #100medium of 61033.6
  2. #105high of 61032.6
  3. #178minimal of 61023.8

Leader: Opus 5.5 · 57.6

composite

Arena Text

220 results
  1. #27high of 2201478
  2. #29medium of 2201476

Leader: Gemini 4 · 1525

composite

Arena Text — Multi-Turn

220 results
  1. #39high of 2201480
  2. #43medium of 2201478

Leader: Gemini 4 · 1553

composite

Vals Index

47 results
  1. #38high of 4744.8%

Leader: Gemini 4 · 68.9%

composite

Vals Multimodal Index v1.2

33 results
  1. #16high of 3362.9%

Leader: Fable 5 · 74.2%

cost

AA Intelligence Index Cost per Task

177 results
  1. #164high of 177$3.69

Leader: GPT-6 Luna · $0.00

cost

Cost to Run AA Intelligence Index

177 results
  1. #160high of 177$4,222

Leader: GPT-6 Luna · $10.63

creative

MineBench

72 results
  1. #26xhigh of 721687

Leader: GPT-6 Astra Pro · 2281

creative

VoxelBench (image)

16 results
  1. #16xhigh of 161000

Leader: GPT-6.1 Sol · 2204

creative

VoxelBench (text)

55 results
  1. #24xhigh of 551700

Leader: GPT-6 Astra · 2733

economics

CorpFin (Vals)

117 results
  1. #36high of 11764.7%

Leader: Opus 5 · 73.2%

economics

Excel Modeling Benchmark

74 results
  1. #27high of 7463.5%

Leader: Fable 5.1 · 76.7%

economics

Finance Agent v2

78 results
  1. #12high of 7857.9%

Leader: Gemini 4 · 65.4%

economics

GDP.pdf

210 results
  1. #65high of 21019.8%

Leader: Muse Wonder Full (Max) · 32.6%

economics

GDPval-AA v2.1

288 results
  1. #115high of 2881202

Leader: Opus 5.5 · 1866

economics

MortgageTax (Vals)

84 results
  1. #19high of 8468.1%

Leader: Opus 5 · 72.1%

economics

Public Benefits Bench v1 (Vals)

13 results
  1. #5high of 1358.0%

Leader: Fable 5 · 71.7%

economics

TaxEval (Vals)

125 results
  1. #37high of 12574.4%

Leader: Muse Spark 1.2 · 80.4%

instruction

IFBench

399 results
  1. #21high of 39976.3%
  2. #35medium of 39974.6%
  3. #187minimal of 39947.3%

Leader: Grok 4.3 · 83.3%

knowledge

AA-Omniscience Accuracy

526 results
  1. #56high of 52651.4%
  2. #58medium of 52651.0%
  3. #100minimal of 52643.0%

Leader: Fable 5.1 · 67.2%

knowledge

AA-Omniscience Non-hallucination

526 results
  1. #166medium of 52638.2%
  2. #167high of 52637.8%
  3. #233minimal of 52625.7%

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

knowledge

GPQA Diamond

539 results
  1. #38high of 53992.2%
  2. #40medium of 53992.1%
  3. #165minimal of 53982.8%

Leader: GPT-6 Astra · 96.3%

knowledge

GPQA Diamond (Vals)

122 results
  1. #14high of 12292.7%

Leader: Gemini 3.1 Pro · 95.5%

knowledge

MedCode

94 results
  1. #6high of 9455.8%

Leader: Opus 5 · 63.6%

knowledge

MedScribe

96 results
  1. #66high of 9676.6%

Leader: Opus 5.5 · 91.4%

knowledge

MMLU Pro (Vals)

122 results
  1. #10high of 12289.5%

Leader: Fable 5.1 · 92.4%

knowledge

MMMU Pro (Vals)

81 results
  1. #8high of 8188.3%

Leader: Fable 5.1 · 90.6%

knowledge

MMMU-Pro

265 results
  1. #20high of 26584.3%
  2. #24medium of 26583.9%
  3. #50minimal of 26580.1%

Leader: Opus 5.5 · 87.7%

knowledge

Public Benefits Bench v1.1

50 results
  1. #33high of 5059.5%

Leader: Opus 5 · 76.9%

long-context

AA-LCR v1.1

523 results
  1. #161medium of 52374.3%
  2. #177high of 52373.3%
  3. #263minimal of 52361.3%

Leader: Kimi K3 · 88.7%

long-context

Arena Document

43 results
  1. #20medium of 431463
  2. #22high of 431461

Leader: Opus 5 · 1516

long-context

MLCR-AA

94 results
  1. #36high of 9418.3%

Leader: Sonnet 5.5 · 75.0%

math

FrontierMath Tiers 1–3 (v2)

115 results
  1. #35high of 11562.8%

Leader: GPT-6.1 Sol · 93.7%

math

ProofBench v1.1

52 results
  1. #37high of 5231.0%

Leader: Sonnet 5.5 · 100.0%

other

Arena Vision

116 results
  1. #21medium of 1161285
  2. #23high of 1161284

Leader: Fable 5 · 1308

other

T3 Code usage share

85 results
  1. #34xhigh of 850.0%

Leader: Opus 5.5 · 29.1%

reasoning

ARC-AGI-1

215 results
  1. #49high of 21592.5%
  2. #153minimal of 21548.8%

Leader: Fable 5 · 98.5%

reasoning

ARC-AGI-2

216 results
  1. #49high of 21672.1%
  2. #141minimal of 2168.9%

Leader: GPT-6 Astra · 95.0%

reasoning

Blueprint-Bench 2

28 results
  1. #10xhigh of 2833.6%

Leader: Gemini 4 · 54.4%

reasoning

CritPt

529 results
  1. #98high of 52913.1%
  2. #110medium of 52910.9%
  3. #204minimal of 5291.4%

Leader: Muse Wonder Full (Max) · 32.9%

reasoning

Humanity's Sixth Sense

25 results
  1. #10high of 2532.6%

Leader: GPT-6 Astra · 53.6%

reasoning

LegalBench

135 results
  1. #51high of 13583.6%

Leader: Fable 5 · 88.6%

reasoning

Roboflow Visual Reasoning

102 results
  1. #9low of 10282.1%
  2. #12high of 10281.0%

Leader: GPT-6 Astra · 91.2%

reasoning

SAGE

80 results
  1. #18high of 8049.9%

Leader: Opus 4.7 · 56.1%

reasoning

SimpleQA Verified

47 results
  1. #7xhigh of 4770.4%

Leader: Gemini 3.1 Pro · 77.5%

reliability

Tool-call error rate

172 results
  1. #94xhigh of 1721.5%

Leader: Gemini 4 · 0.0%

reliability

Uptime (7d)

182 results
  1. #151xhigh of 18298.0%

Leader: GPT-5 Pro · 100.0%

speed

Latency (Time to First Token)

351 results
  1. #32minimal of 3510.8s
  2. #258medium of 35113s
  3. #270high of 35117s

Leader: Gemini 2.5 Flash Lite · 0.3s

speed

Output Speed (tokens/s)

351 results
  1. #41medium of 351201.9
  2. #44high of 351197.4
  3. #52minimal of 351189.3

Leader: Celeris-1 · 1564.1

speed

Provider latency

178 results
  1. #103xhigh of 1781.5s

Leader: Qwen 2.5 7B · 0.2s

speed

Provider throughput

176 results
  1. #17xhigh of 176116.0

Leader: Muse Spark 1.1 · 204.5