priorsBuilder

Anthropic · released 2026-04-16

Opus 4.7

Measured on 90 benchmarks across 5 reasoning efforts.

General frontier

56.1

#12 of 32 models · best at none

Reasoning efforts in your mix

none56.1#40 of 128
xhigh43.5#75 of 128
max43.5#76 of 128
high37.5#96 of 128
medium36.0#105 of 128

Your mix’s benchmarks

6 of 8 measured

reasoning

Humanity's Last Exam

587 results
  1. #67max of 58742.3%
  2. #133high of 58733.3%

Leader: Opus 5.5 · 61.4%

knowledge

AA-Omniscience

526 results
  1. #41max of 52627.3
  2. #80high of 52614.8

Leader: Opus 5.5 · 46.4

reasoning

ARC-AGI-3

57 results
  1. #48high of 570.2%

Leader: GPT-6 Astra · 99.9%

reasoning

BullshitBench v2

196 results
  1. #9none of 19683.0%
  2. #20max of 19674.0%

Leader: Opus 4.8 · 95.0%

math

FrontierMath Tier 4 (v2)

70 results
  1. #33max of 7031.7%

Leader: GPT-6.1 Sol · 100.0%

reasoning

SimpleBench

97 results
  1. #31max of 9761.7%

Leader: Opus 5.5 · 88.4%

All 90 benchmarks

show 84 more

agentic

AA-AnalystAgent

44 results
  1. #16max of 4443.8%

Leader: Gemini 3.7 Flash · 60.0%

agentic

AA-Briefcase v1.1

224 results
  1. #81max of 2241254
  2. #107high of 2241068

Leader: Sonnet 5.5 · 1823

agentic

AutomationBench (Zapier)

115 results
  1. #60max of 11513.4%
  2. #73xhigh of 1159.7%
  3. #76high of 1159.1%
  4. #86medium of 1156.8%
  5. #89low of 1155.5%

Leader: Gemini 4 · 51.3%

agentic

ExploitBench v0.1

4 results
  1. #4max of 428.0%

Leader: Mythos Preview · 78.0%

agentic

Harvey's Legal Agent Benchmark

79 results
  1. #30max of 796.7%

Leader: Muse Spark 1.2 · 25.4%

agentic

ITBench-AA

45 results
  1. #11max of 4546.7%

Leader: GPT-5.6 Sol · 56.2%

agentic

Legal Research Bench

78 results
  1. #33max of 7838.5%

Leader: Muse Spark 1.3 · 55.3%

agentic

PostTrainBench v1.1

12 results
  1. #8max of 1228.6%

Leader: Fable 5 · 41.8%

agentic

Tax Agent Bench

69 results
  1. #38max of 6923.7%

Leader: Fable 5.1 · 49.2%

agentic

Terminal-Bench 2.0

40 results
  1. #3max of 4080.2%

Leader: GPT-5.5 · 84.7%

agentic

Terminal-Bench 2.1

18 results
  1. #16max of 1868.9%

Leader: GPT-6 Astra · 87.4%

agentic

Vending-Bench 2

67 results
  1. #6max of 67$10,937

Leader: GPT-6 Astra · $15,515

agentic

Vending-Bench Arena

12 results
  1. #7max of 12$5,838

Leader: GPT-6 Astra · $12,400

agentic

τ²-Bench Telecom

393 results
  1. #73max of 39388.6%
  2. #146high of 39374.0%

Leader: GLM 5.2 · 99.1%

agentic

τ³-Banking

206 results
  1. #49max of 20634.6%

Leader: Qwen 3.8 Max · 51.3%

coding

Arena Image-to-WebDev

58 results
  1. #21high of 581578
  2. #26max of 581562

Leader: Opus 5.5 · 1749

coding

Arena WebDev

122 results
  1. #34high of 1221557
  2. #36max of 1221555

Leader: Opus 5.5 · 1813

coding

Code Migration

77 results
  1. #25max of 7743.9%

Leader: Sonnet 5.5 · 69.8%

coding

DeepSWE v1

29 results
  1. #6max of 2954.2%
  2. #10xhigh of 2944.7%
  3. #12high of 2940.3%
  4. #14medium of 2931.6%

Leader: GPT-5.5 · 70.0%

coding

FrontierCode 1.1 Extended

130 results
  1. #72max of 13053.9%
  2. #80xhigh of 13051.4%
  3. #84high of 13051.1%
  4. #100medium of 13044.8%
  5. #104low of 13043.6%

Leader: Opus 5.5 · 65.3%

coding

FrontierCode 1.1 Main

130 results
  1. #76max of 13038.5%
  2. #90high of 13035.2%
  3. #92xhigh of 13034.9%
  4. #101medium of 13028.9%
  5. #104low of 13027.6%

Leader: Opus 5.5 · 54.6%

coding

FrontierSWE v1

17 results
  1. #8max of 1756.3%

Leader: Fable 5 · 88.2%

coding

LiveCodeBench (Vals)

127 results
  1. #36max of 12785.1%

Leader: Fable 5.1 · 90.5%

coding

ProgramBench v1 Almost Resolved

62 results
  1. #27max of 629.5%

Leader: Opus 5.5 · 65.0%

coding

ProgramBench v1 Fully Resolved

62 results
  1. #34max of 620.0%

Leader: Opus 5.5 · 18.5%

coding

ProgramBench v1 Raw Pass Rate

62 results
  1. #31max of 6264.2%

Leader: Opus 5.5 · 87.0%

coding

SWE Atlas-QnA

22 results
  1. #11max of 2240.3%

Leader: Opus 5 · 63.2%

coding

SWE-bench Verified (Vals)

85 results
  1. #21max of 8582.0%

Leader: Opus 5 · 97.0%

coding

Terminal-Bench 2.0 (Vals)

63 results
  1. #3high of 6368.5%

Leader: GPT-5.5 · 73.2%

coding

Terminal-Bench 2.1 (Vals)

74 results
  1. #30high of 7468.5%

Leader: Opus 5.5 · 87.6%

coding

Terminal-Bench Hard-AA

387 results
  1. #15high of 38754.5%
  2. #21max of 38751.5%

Leader: GPT-5.6 Sol · 65.9%

coding

Terminal-Bench v2.1-AA

238 results
  1. #36max of 23883.1%

Leader: Fable 5.1 · 91.4%

coding

Vibe Code Bench v1.1

105 results
  1. #36max of 10571.0%

Leader: Sonnet 5.5 · 92.4%

coding

WeirdML v2

127 results
  1. #22max of 12776.4%
  2. #24xhigh of 12775.4%

Leader: GPT-6 Astra Pro · 93.6%

composite

AA Intelligence Index v4.3.2

610 results
  1. #59max of 61040.7
  2. #112high of 61030.9

Leader: Opus 5.5 · 57.6

composite

Arena Text

220 results
  1. #5high of 2201501
  2. #10max of 2201494

Leader: Gemini 4 · 1525

composite

Arena Text — Multi-Turn

220 results
  1. #4high of 2201516
  2. #5max of 2201514

Leader: Gemini 4 · 1553

composite

Vals Multimodal Index v1.2

33 results
  1. #10high of 3367.4%

Leader: Fable 5 · 74.2%

composite

Vals RSI Index v1.1

24 results
  1. #18max of 2419.1%

Leader: Opus 5.5 · 37.3%

composite

Vals RSI Index v1.1: Harness Engineering: Judge

24 results
  1. #21max of 245.6%

Leader: GPT-6 Sol · 22.8%

composite

Vals RSI Index v1.1: Post-training: Finance Agent

24 results
  1. #24max of 240.0%

Leader: Opus 5.5 · 35.1%

composite

Vals RSI Index v1.1: Pre-training: Compression

24 results
  1. #15max of 2431.4%

Leader: Opus 5.5 · 46.5%

composite

Vals RSI Index v1.1: Pre-training: LM Training

24 results
  1. #15max of 2439.3%

Leader: Opus 5.5 · 53.0%

creative

MineBench

72 results
  1. #29max of 721643

Leader: GPT-6 Astra Pro · 2281

creative

VoxelBench (image)

16 results
  1. #9max of 161276

Leader: GPT-6.1 Sol · 2214

creative

VoxelBench (text)

55 results
  1. #41max of 551358

Leader: GPT-6 Astra · 2732

economics

CorpFin (Vals)

117 results
  1. #23max of 11766.1%

Leader: Opus 5 · 73.2%

economics

Excel Modeling Benchmark

74 results
  1. #26max of 7463.7%

Leader: Fable 5.1 · 76.7%

economics

Finance Agent v2

78 results
  1. #38high of 7851.5%

Leader: Gemini 4 · 65.4%

economics

GDPval-AA v2.1

288 results
  1. #85max of 2881353

Leader: Opus 5.5 · 1866

economics

MortgageTax (Vals)

84 results
  1. #3max of 8470.3%

Leader: Opus 5 · 72.1%

economics

TaxEval (Vals)

125 results
  1. #21max of 12575.3%

Leader: Muse Spark 1.2 · 80.4%

instruction

IFBench

399 results
  1. #128max of 39958.6%
  2. #220high of 39943.6%

Leader: Grok 4.3 · 83.3%

knowledge

AA-Omniscience Accuracy

526 results
  1. #66max of 52648.9%
  2. #90high of 52644.7%

Leader: Fable 5.1 · 67.2%

knowledge

AA-Omniscience Non-hallucination

526 results
  1. #82max of 52657.7%
  2. #135high of 52645.9%

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

knowledge

Arena Search

28 results
  1. #4max of 281233

Leader: GPT-5.6 Sol · 1257

knowledge

CaseLaw (Vals)

53 results
  1. #5max of 5368.4%

Leader: Grok 4.3 · 79.3%

knowledge

GPQA Diamond

539 results
  1. #47max of 53991.4%
  2. #89high of 53988.5%

Leader: GPT-6 Astra · 96.3%

knowledge

GPQA Diamond (Vals)

122 results
  1. #26max of 12290.2%

Leader: Gemini 3.1 Pro · 95.5%

knowledge

MedCode

94 results
  1. #7max of 9454.9%

Leader: Opus 5 · 63.6%

knowledge

MedScribe

96 results
  1. #45max of 9683.0%

Leader: Opus 5.5 · 91.4%

knowledge

MMLU Pro (Vals)

122 results
  1. #8max of 12289.9%

Leader: Fable 5.1 · 92.4%

knowledge

MMMU Pro (Vals)

81 results
  1. #24max of 8185.5%

Leader: Fable 5.1 · 90.6%

knowledge

MMMU-Pro

265 results
  1. #64max of 26578.8%
  2. #86high of 26576.4%

Leader: Opus 5.5 · 87.7%

long-context

AA-LCR v1.1

523 results
  1. #107max of 52378.7%
  2. #150high of 52375.7%

Leader: Kimi K3 · 88.7%

long-context

Arena Document

43 results
  1. #6max of 431495
  2. #7high of 431495

Leader: Opus 5 · 1516

long-context

MLCR-AA

94 results
  1. #21high of 9428.9%

Leader: Sonnet 5.5 · 75.0%

math

AIME (Vals)

82 results
  1. #7max of 8296.3%

Leader: Gemini 3.1 Pro · 98.1%

math

FrontierMath Tiers 1–3 (v2)

115 results
  1. #24max of 11570.2%

Leader: GPT-6.1 Sol · 93.7%

other

Arena Vision

116 results
  1. #4max of 1161298
  2. #5high of 1161298

Leader: Fable 5 · 1308

other

T3 Code usage share

85 results
  1. #20max of 850.0%

Leader: Opus 5.5 · 29.1%

reasoning

ARC-AGI-1

215 results
  1. #43high of 21593.5%
  2. #55max of 21592.0%
  3. #60low of 21591.0%
  4. #61medium of 21591.0%

Leader: Fable 5 · 98.5%

reasoning

ARC-AGI-2

216 results
  1. #44max of 21675.8%
  2. #56high of 21668.3%
  3. #58medium of 21667.5%
  4. #70low of 21662.1%

Leader: GPT-6 Astra · 95.0%

reasoning

Blueprint-Bench 2

28 results
  1. #24max of 2824.5%

Leader: Gemini 4 · 54.4%

reasoning

CritPt

529 results
  1. #103max of 52912.0%
  2. #149high of 5295.1%

Leader: Muse Wonder Full (Max) · 32.9%

reasoning

Furniture Assembly

31 results
  1. #21max of 3133.3%

Leader: Opus 5.5 · 83.3%

reasoning

LegalBench

135 results
  1. #23max of 13585.3%

Leader: Fable 5 · 88.6%

reasoning

SAGE

80 results
  1. #1max of 8056.1%

Leader: this model · 56.1%

reliability

Tool-call error rate

171 results
  1. #75max of 1711.0%

Leader: GPT-5.4 Pro · 0.0%

reliability

Uptime (7d)

181 results
  1. #38max of 18199.9%

Leader: GPT-5 Pro · 100.0%

speed

Latency (Time to First Token)

351 results
  1. #107high of 3511.5s
  2. #286max of 35122s

Leader: Gemini 2.5 Flash-Lite · 0.3s

speed

Output Speed (tokens/s)

351 results
  1. #294max of 35152.2
  2. #309high of 35143.7

Leader: Celeris-1 · 1522.5

speed

Provider latency

177 results
  1. #131max of 1772.3s

Leader: Qwen 2.5 7B · 0.2s

speed

Provider throughput

176 results
  1. #57max of 17680.3

Leader: Muse Spark 1.1 · 204.3