priorsBuilder

Anthropic · released 2026-02-17

Sonnet 4.6

Measured on 87 benchmarks.

General frontier

Not ranked in this mix — too little measured data for its benchmarks.

Your mix’s benchmarks

4 of 8 measured

reasoning

Humanity's Last Exam

587 results
  1. #130max of 58733.6%
  2. #262none of 58713.3%
  3. #287low of 58711.2%

Leader: Opus 5.5 · 61.4%

knowledge

AA-Omniscience

526 results
  1. #87max of 52612.2
  2. #143low of 526-2.1
  3. #149none of 526-3.5

Leader: Opus 5.5 · 46.4

reasoning

BullshitBench v2

196 results
  1. #4high of 19691.0%
  2. #6none of 19689.0%

Leader: Opus 4.8 · 95.0%

coding

DeepSWE v1.1

70 results
  1. #65high of 7029.9%

Leader: GPT-6 Astra · 74.1%

All 87 benchmarks

show 83 more

agentic

AA-AnalystAgent

44 results
  1. #28max of 4420.0%

Leader: Gemini 3.7 Flash · 60.0%

agentic

AA-Briefcase v1.1

224 results
  1. #108max of 2241063

Leader: Sonnet 5.5 · 1823

agentic

APEX-Agents-AA

33 results
  1. #15max of 3328.0%

Leader: Gemini 3.5 Flash · 47.1%

agentic

AutomationBench (Zapier)

115 results
  1. #91max of 1155.3%
  2. #92high of 1155.0%
  3. #100medium of 1153.6%
  4. #103low of 1152.9%

Leader: Gemini 4 · 51.3%

agentic

AutomationBench-AA

219 results
  1. #132max of 21920.1%

Leader: Gemini 4 · 77.5%

agentic

Harvey's Legal Agent Benchmark

79 results
  1. #36max of 795.0%

Leader: Muse Spark 1.2 · 25.4%

agentic

ITBench-AA

45 results
  1. #21max of 4539.8%

Leader: GPT-5.6 Sol · 56.2%

agentic

Legal Research Bench

78 results
  1. #34max of 7838.5%

Leader: Muse Spark 1.3 · 55.3%

agentic

OSWorld-Verified

13 results
  1. #7max of 1372.1%

Leader: Fable 5 · 86.0%

agentic

SkillsBench

34 results
  1. #23max of 3449.1%

Leader: DeepSeek V4.1 Flash · 69.8%

agentic

Tax Agent Bench

69 results
  1. #34max of 6925.8%

Leader: Fable 5.1 · 49.2%

agentic

Terminal-Bench 2.0

40 results
  1. #14max of 4053.4%

Leader: GPT-5.5 · 84.7%

agentic

Vending-Bench 2

67 results
  1. #16max of 67$7,204

Leader: GPT-6 Astra · $15,515

agentic

Vending-Bench Arena

12 results
  1. #8max of 12$5,639

Leader: GPT-6 Astra · $12,400

agentic

τ²-Bench Telecom

393 results
  1. #127none of 39379.5%
  2. #130low of 39378.9%
  3. #138max of 39375.7%

Leader: GLM 5.2 · 99.1%

agentic

τ³-Banking

206 results
  1. #51max of 20634.4%

Leader: Qwen 3.8 Max · 51.3%

coding

Arena Image-to-WebDev

58 results
  1. #34max of 581529

Leader: Opus 5.5 · 1749

coding

Arena WebDev

122 results
  1. #46max of 1221522

Leader: Opus 5.5 · 1813

coding

Code Migration

77 results
  1. #30max of 7739.9%

Leader: Sonnet 5.5 · 69.8%

coding

DeepSWE v1

29 results
  1. #13high of 2931.8%

Leader: GPT-5.5 · 70.0%

coding

FrontierCode 1.1 Extended

130 results
  1. #110max of 13040.0%
  2. #111high of 13039.9%
  3. #115medium of 13037.5%
  4. #122low of 13030.4%

Leader: Opus 5.5 · 65.3%

coding

FrontierCode 1.1 Main

130 results
  1. #111max of 13024.3%
  2. #114high of 13023.5%
  3. #117medium of 13021.1%
  4. #123low of 13015.0%

Leader: Opus 5.5 · 54.6%

coding

LiveCodeBench (Vals)

127 results
  1. #57max of 12782.1%

Leader: Fable 5.1 · 90.5%

coding

ProgramBench v1 Almost Resolved

62 results
  1. #42max of 622.0%

Leader: Opus 5.5 · 65.0%

coding

ProgramBench v1 Fully Resolved

62 results
  1. #17max of 620.5%

Leader: Opus 5.5 · 18.5%

coding

ProgramBench v1 Raw Pass Rate

62 results
  1. #37max of 6259.1%

Leader: Opus 5.5 · 87.0%

coding

SciCode

226 results
  1. #106max of 22650.1%

Leader: Opus 5.5 · 66.9%

coding

SWE Atlas-QnA

22 results
  1. #15max of 2231.2%

Leader: Opus 5 · 63.2%

coding

SWE-bench Verified (Vals)

85 results
  1. #34max of 8577.4%

Leader: Opus 5 · 97.0%

coding

Terminal-Bench 2.0 (Vals)

63 results
  1. #8max of 6359.6%

Leader: GPT-5.5 · 73.2%

coding

Terminal-Bench 2.1 (Vals)

74 results
  1. #44max of 7457.3%

Leader: Opus 5.5 · 87.6%

coding

Terminal-Bench Hard-AA

387 results
  1. #18max of 38753.0%
  2. #32none of 38746.2%
  3. #47low of 38742.4%

Leader: GPT-5.6 Sol · 65.9%

coding

Terminal-Bench v2.1-AA

238 results
  1. #70max of 23871.2%

Leader: Fable 5.1 · 91.4%

coding

Terminal-Bench v4-AA

216 results
  1. #106max of 2163.0%

Leader: Sonnet 5.5 · 63.6%

coding

Vibe Code Bench v1.1

105 results
  1. #48max of 10555.8%

Leader: Sonnet 5.5 · 92.4%

coding

WeirdML v2

127 results
  1. #36max of 12766.1%

Leader: GPT-6 Astra Pro · 93.6%

composite

AA Intelligence Index v4.3.2

610 results
  1. #120max of 61030.1
  2. #168none of 61024.7
  3. #184low of 61023.3

Leader: Opus 5.5 · 57.6

composite

Arena Text

220 results
  1. #43max of 2201472

Leader: Gemini 4 · 1525

composite

Arena Text — Multi-Turn

220 results
  1. #37max of 2201481

Leader: Gemini 4 · 1553

composite

Vals Multimodal Index v1.2

33 results
  1. #17max of 3360.6%

Leader: Fable 5 · 74.2%

cost

AA Intelligence Index Cost per Task

177 results
  1. #155max of 177$2.49

Leader: GPT-6 Luna · $0.00

cost

Cost to Run AA Intelligence Index

177 results
  1. #162max of 177$4,925

Leader: GPT-6 Luna · $10.63

creative

MineBench

72 results
  1. #43max of 721393

Leader: GPT-6 Astra Pro · 2281

economics

CorpFin (Vals)

117 results
  1. #31max of 11765.3%

Leader: Opus 5 · 73.2%

economics

Excel Modeling Benchmark

74 results
  1. #33max of 7460.2%

Leader: Fable 5.1 · 76.7%

economics

Finance Agent v2

78 results
  1. #39max of 7851.0%

Leader: Gemini 4 · 65.4%

economics

GDP.pdf

210 results
  1. #89max of 21015.8%

Leader: Muse Wonder Full (Max) · 32.6%

economics

GDPval-AA v2.1

288 results
  1. #111max of 2881232

Leader: Opus 5.5 · 1866

economics

MortgageTax (Vals)

84 results
  1. #22max of 8467.7%

Leader: Opus 5 · 72.1%

economics

Public Benefits Bench v1 (Vals)

13 results
  1. #4max of 1358.5%

Leader: Fable 5 · 71.7%

economics

TaxEval (Vals)

125 results
  1. #4max of 12577.1%

Leader: Muse Spark 1.2 · 80.4%

instruction

IFBench

399 results
  1. #141max of 39956.6%
  2. #235low of 39942.4%
  3. #247none of 39941.2%

Leader: Grok 4.3 · 83.3%

knowledge

AA-Omniscience Accuracy

526 results
  1. #116max of 52640.9%
  2. #132none of 52638.6%
  3. #145low of 52636.5%

Leader: Fable 5.1 · 67.2%

knowledge

AA-Omniscience Non-hallucination

526 results
  1. #105max of 52651.6%
  2. #159low of 52639.2%
  3. #207none of 52631.5%

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

knowledge

Arena Search

28 results
  1. #6max of 281221

Leader: GPT-5.6 Sol · 1257

knowledge

CaseLaw (Vals)

53 results
  1. #14max of 5364.0%

Leader: Grok 4.3 · 79.3%

knowledge

Global-MMLU-Lite

118 results
  1. #11max of 11890.5%
  2. #21none of 11889.2%
  3. #25low of 11888.2%

Leader: Gemini 3.1 Pro · 93.2%

knowledge

GPQA Diamond

539 results
  1. #96max of 53987.5%
  2. #193none of 53979.9%
  3. #194low of 53979.7%

Leader: GPT-6 Astra · 96.3%

knowledge

GPQA Diamond (Vals)

122 results
  1. #48max of 12285.6%

Leader: Gemini 3.1 Pro · 95.5%

knowledge

MedQA (Vals)

80 results
  1. #34max of 8092.1%

Leader: o1 · 96.5%

knowledge

MMLU Pro (Vals)

122 results
  1. #29max of 12287.3%

Leader: Fable 5.1 · 92.4%

knowledge

MMMU Pro (Vals)

81 results
  1. #31max of 8183.6%

Leader: Fable 5.1 · 90.6%

knowledge

MMMU-Pro

265 results
  1. #120max of 26573.3%
  2. #137none of 26570.6%
  3. #152low of 26569.2%

Leader: Opus 5.5 · 87.7%

knowledge

Public Benefits Bench v1.1

50 results
  1. #27max of 5062.4%

Leader: Opus 5 · 76.9%

long-context

AA-LCR v1.1

523 results
  1. #83max of 52380.0%
  2. #220low of 52369.3%
  3. #228none of 52368.3%

Leader: Kimi K3 · 88.7%

long-context

Arena Document

43 results
  1. #11max of 431482

Leader: Opus 5 · 1516

long-context

MLCR-AA

94 results
  1. #23max of 9424.4%
  2. #29low of 9420.6%
  3. #31none of 9420.0%

Leader: Sonnet 5.5 · 75.0%

math

AIME (Vals)

82 results
  1. #20max of 8292.3%

Leader: Gemini 3.1 Pro · 98.1%

other

Arena Vision

116 results
  1. #33max of 1161275

Leader: Fable 5 · 1308

other

T3 Code usage share

85 results
  1. #17max of 850.1%

Leader: Opus 5.5 · 29.1%

reasoning

ARC-AGI-1

215 results
  1. #90high of 21586.5%
  2. #94max of 21586.0%

Leader: Fable 5 · 98.5%

reasoning

ARC-AGI-2

216 results
  1. #76high of 21660.4%
  2. #85max of 21658.3%

Leader: GPT-6 Astra · 95.0%

reasoning

Blueprint-Bench 2

28 results
  1. #27max of 286.7%

Leader: Gemini 4 · 54.4%

reasoning

CritPt

529 results
  1. #172max of 5293.1%
  2. #231none of 5290.9%
  3. #232low of 5290.9%

Leader: Muse Wonder Full (Max) · 32.9%

reasoning

LegalBench

135 results
  1. #71max of 13582.1%

Leader: Fable 5 · 88.6%

reasoning

SAGE

80 results
  1. #34max of 8046.6%

Leader: Opus 4.7 · 56.1%

reasoning

SimpleQA Verified

47 results
  1. #28max of 4730.9%

Leader: Gemini 3.1 Pro · 77.5%

reliability

Tool-call error rate

171 results
  1. #92max of 1711.4%

Leader: GPT-5.4 Pro · 0.0%

reliability

Uptime (7d)

181 results
  1. #142max of 18198.3%

Leader: GPT-5 Pro · 100.0%

speed

Latency (Time to First Token)

351 results
  1. #62low of 3511.0s
  2. #69none of 3511.0s
  3. #329max of 351103s

Leader: Gemini 2.5 Flash-Lite · 0.3s

speed

Output Speed (tokens/s)

351 results
  1. #264max of 35158.8
  2. #306none of 35145.2
  3. #311low of 35143.6

Leader: Celeris-1 · 1522.5

speed

Provider latency

177 results
  1. #60max of 1771.0s

Leader: Qwen 2.5 7B · 0.2s

speed

Provider throughput

176 results
  1. #86max of 17667.0

Leader: Muse Spark 1.1 · 204.3