priorsBuilder

Anthropic · released 2026-09-01

Fable 5.1

Measured on 99 benchmarks across 5 reasoning efforts.

General frontier

72.6

#4 of 32 models · best at xhigh

Reasoning efforts in your mix

xhigh72.6#8 of 128
max72.5#9 of 128
high71.5#14 of 128
medium69.6#16 of 128
low63.7#27 of 128

Your mix’s benchmarks

6 of 8 measured

reasoning

Humanity's Last Exam

584 results
  1. #2max of 58459.1%
  2. #3xhigh of 58458.7%
  3. #6high of 58455.9%
  4. #15medium of 58453.8%
  5. #28low of 58448.9%

Leader: Opus 5.5 · 61.4%

knowledge

AA-Omniscience

523 results
  1. #3max of 52343.5
  2. #8xhigh of 52342.4
  3. #14high of 52340.8
  4. #21medium of 52337.6
  5. #24low of 52334.1

Leader: Opus 5.5 · 46.4

reasoning

BullshitBench v2

192 results
  1. #16low of 19277.0%
  2. #36max of 19264.0%

Leader: Opus 4.8 · 95.0%

math

FrontierMath Tier 4 (v2)

70 results
  1. #9max of 7087.8%

Leader: GPT-6.1 Sol · 100.0%

reasoning

SimpleBench

97 results
  1. #2max of 9786.6%

Leader: Opus 5.5 · 88.4%

agentic

Terminal-Bench 4.0

35 results
  1. #5max of 3557.9%
  2. #8xhigh of 3557.9%
  3. #9high of 3554.5%
  4. #11medium of 3553.9%
  5. #19low of 3543.3%

Leader: Opus 5.5 · 64.8%

All 99 benchmarks

show 93 more

agentic

AA-AnalystAgent

39 results
  1. #3max of 3957.5%

Leader: Gemini 3.7 Flash · 60.0%

agentic

AA-Briefcase v1.1

222 results
  1. #6max of 2221675
  2. #8xhigh of 2221656
  3. #17high of 2221581
  4. #27medium of 2221528
  5. #38low of 2221482

Leader: Sonnet 5.5 · 1823

agentic

Agent Arena

49 results
  1. #3max of 4912.7%

Leader: Opus 5.5 · 14.3%

agentic

AutomationBench (Zapier)

115 results
  1. #33max of 11522.4%
  2. #34xhigh of 11521.3%

Leader: Gemini 4 · 51.3%

agentic

AutomationBench-AA

217 results
  1. #36max of 21759.4%
  2. #43xhigh of 21757.8%
  3. #52high of 21755.3%
  4. #56medium of 21754.7%
  5. #68low of 21752.2%

Leader: Gemini 4 · 77.5%

agentic

CUA-bench

8 results
  1. #3max of 813.2%

Leader: GPT-6 Astra · 19.2%

agentic

Harvey's Legal Agent Benchmark

76 results
  1. #30max of 766.7%

Leader: Muse Spark 1.2 · 25.4%

agentic

ITBench-AA

45 results
  1. #6max of 4549.5%

Leader: GPT-5.6 Sol · 56.2%

agentic

Legal Research Bench

75 results
  1. #3max of 7555.3%

Leader: Muse Spark 1.3 · 55.3%

agentic

SkillsBench

34 results
  1. #5max of 3461.6%

Leader: DeepSeek V4.1 Flash · 69.8%

agentic

Tax Agent Bench

67 results
  1. #1max of 6749.2%

Leader: this model · 49.2%

agentic

Terminal-Bench 4.0 (Vals)

45 results
  1. #4max of 4558.1%

Leader: Opus 5.5 · 65.2%

agentic

Terminal-Bench Science (Vals)

39 results
  1. #6max of 3940.0%

Leader: GPT-6 Astra · 62.9%

agentic

Time Horizon Index: KSP

13 results
  1. #3max of 1363.3%

Leader: Opus 5.5 · 91.3%

agentic

Vending-Bench 2

67 results
  1. #26max of 67$5,422

Leader: GPT-6 Astra · $15,515

agentic

WeirdML v3

17 results
  1. #4xhigh of 1726.0%

Leader: GPT-6 Astra · 42.2%

agentic

τ³-Banking

206 results
  1. #8max of 20647.2%
  2. #12xhigh of 20645.8%
  3. #22high of 20643.1%
  4. #27medium of 20641.0%
  5. #34low of 20639.0%

Leader: Qwen 3.8 Max · 51.3%

coding

Arena Image-to-WebDev

58 results
  1. #4max of 581720

Leader: Opus 5.5 · 1749

coding

Arena WebDev

122 results
  1. #5max of 1221744

Leader: Opus 5.5 · 1813

coding

Code Migration

75 results
  1. #9max of 7554.6%

Leader: Sonnet 5.5 · 69.8%

coding

Drone-Bench

11 results
  1. #3max of 1191.2%

Leader: Opus 5.5 · 97.3%

coding

FrontierCode 1.1 Extended

130 results
  1. #9medium of 13063.6%
  2. #14high of 13062.7%
  3. #17max of 13062.0%
  4. #18low of 13061.6%
  5. #20xhigh of 13061.4%

Leader: Opus 5.5 · 65.3%

coding

FrontierCode 1.1 Main

130 results
  1. #12medium of 13050.9%
  2. #14high of 13050.3%
  3. #15max of 13050.3%
  4. #19low of 13049.8%
  5. #24xhigh of 13048.7%

Leader: Opus 5.5 · 54.6%

coding

FrontierSWE v2

21 results
  1. #4max of 2156.3%

Leader: GPT-6 Astra · 65.5%

coding

IOI

42 results
  1. #6max of 4290.8%

Leader: Gemini 4 · 100.0%

coding

LiveCodeBench (Vals)

127 results
  1. #1max of 12790.5%

Leader: this model · 90.5%

coding

ProgramBench v1 Almost Resolved

62 results
  1. #2max of 6253.5%

Leader: Opus 5.5 · 65.0%

coding

ProgramBench v1 Fully Resolved

62 results
  1. #2max of 627.0%

Leader: Opus 5.5 · 18.5%

coding

ProgramBench v1 Raw Pass Rate

62 results
  1. #5max of 6282.7%

Leader: Opus 5.5 · 87.0%

coding

SciCode

224 results
  1. #3max of 22463.1%
  2. #7xhigh of 22460.9%
  3. #18high of 22458.7%
  4. #31low of 22456.7%
  5. #36medium of 22456.4%

Leader: Opus 5.5 · 66.9%

coding

SWE Atlas-QnA

22 results
  1. #2xhigh of 2260.0%

Leader: Opus 5 · 63.2%

coding

Terminal-Bench 2.1 (Vals)

74 results
  1. #4high of 7485.0%

Leader: Opus 5.5 · 87.6%

coding

Terminal-Bench Science-AA

45 results
  1. #8max of 4543.3%

Leader: GPT-6 Astra · 63.3%

coding

Terminal-Bench v2.1-AA

238 results
  1. #1max of 23891.4%
  2. #2xhigh of 23891.0%
  3. #3high of 23889.9%
  4. #13medium of 23888.0%
  5. #26low of 23885.0%

Leader: this model · 91.4%

coding

Terminal-Bench v4-AA

214 results
  1. #10xhigh of 21455.1%
  2. #14max of 21452.0%
  3. #15high of 21452.0%
  4. #22medium of 21444.9%
  5. #28low of 21440.4%

Leader: Sonnet 5.5 · 63.6%

coding

Vibe Code Bench 1-100

22 results
  1. #3max of 2228.0%

Leader: Opus 5.5 · 30.4%

coding

Vibe Code Bench v1.1

103 results
  1. #6max of 10390.3%

Leader: Sonnet 5.5 · 92.4%

coding

WeirdML v2

127 results
  1. #3xhigh of 12792.9%
  2. #5high of 12792.3%

Leader: GPT-6 Astra Pro · 93.6%

composite

AA Intelligence Index v4.3.2

607 results
  1. #5max of 60753.4
  2. #6xhigh of 60753.2
  3. #13high of 60751.2
  4. #21medium of 60748.9
  5. #27low of 60746.8

Leader: Opus 5.5 · 57.6

composite

Arena Text

220 results
  1. #6max of 2201501

Leader: Gemini 4 · 1525

composite

Arena Text — Multi-Turn

220 results
  1. #27max of 2201485

Leader: Gemini 4 · 1553

composite

Vals Index

45 results
  1. #4max of 4565.8%

Leader: Gemini 4 · 68.9%

composite

Vals RSI Index v1.1

24 results
  1. #2max of 2436.1%

Leader: Opus 5.5 · 37.3%

composite

Vals RSI Index v1.1: Harness Engineering: Judge

24 results
  1. #2max of 2422.6%

Leader: GPT-6 Sol · 22.8%

composite

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

24 results
  1. #3max of 2428.0%

Leader: Opus 5.5 · 35.1%

composite

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

24 results
  1. #3max of 2444.3%

Leader: Opus 5.5 · 46.5%

composite

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

24 results
  1. #2max of 2449.5%

Leader: Opus 5.5 · 53.0%

cost

AA Intelligence Index Cost per Task

177 results
  1. #155low of 177$2.37
  2. #161medium of 177$2.98
  3. #167high of 177$3.91
  4. #174xhigh of 177$5.98
  5. #176max of 177$7.63

Leader: GPT-6 Luna · $0.00

cost

Cost to Run AA Intelligence Index

177 results
  1. #152low of 177$3,158
  2. #159medium of 177$3,983
  3. #166high of 177$5,242
  4. #175xhigh of 177$9,063
  5. #177max of 177$13,129

Leader: GPT-6 Luna · $10.63

creative

MineBench

72 results
  1. #8max of 721967

Leader: GPT-6 Astra Pro · 2278

economics

Excel Modeling Benchmark

72 results
  1. #1max of 7276.7%

Leader: this model · 76.7%

economics

Finance Agent v2

76 results
  1. #7max of 7658.9%

Leader: Gemini 4 · 65.4%

economics

GDP.pdf

208 results
  1. #11low of 20828.0%
  2. #16high of 20826.8%
  3. #17medium of 20826.8%
  4. #21max of 20826.2%
  5. #22xhigh of 20826.2%

Leader: GPT-6 Astra · 32.2%

economics

GDPval-AA v2.1

286 results
  1. #4max of 2861756
  2. #5xhigh of 2861733
  3. #18high of 2861633
  4. #41medium of 2861545
  5. #52low of 2861468

Leader: Opus 5.5 · 1866

economics

MortgageTax (Vals)

84 results
  1. #2max of 8470.8%

Leader: Opus 5 · 72.1%

economics

TaxEval (Vals)

125 results
  1. #9max of 12576.0%

Leader: Muse Spark 1.2 · 80.4%

knowledge

AA-Omniscience Accuracy

523 results
  1. #1max of 52367.2%
  2. #3xhigh of 52366.2%
  3. #6high of 52364.9%
  4. #10medium of 52363.1%
  5. #20low of 52360.2%

Leader: this model · 67.2%

knowledge

AA-Omniscience Non-hallucination

523 results
  1. #183low of 52334.4%
  2. #207high of 52331.2%
  3. #210medium of 52330.9%
  4. #216xhigh of 52329.5%
  5. #220max of 52327.4%

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

knowledge

GPQA Diamond

538 results
  1. #11max of 53893.7%
  2. #21xhigh of 53893.4%
  3. #58high of 53890.6%
  4. #88medium of 53888.6%
  5. #92low of 53888.1%

Leader: GPT-6 Astra · 96.3%

knowledge

GPQA Diamond (Vals)

122 results
  1. #9max of 12293.4%

Leader: Gemini 3.1 Pro · 95.5%

knowledge

Harvey LAB-AA v1.1

25 results
  1. #5max of 256.4%

Leader: Grok 4.7 · 9.4%

knowledge

MedCode

94 results
  1. #8max of 9453.5%

Leader: Opus 5 · 63.6%

knowledge

MedScribe

96 results
  1. #2max of 9691.3%

Leader: Opus 5.5 · 91.4%

knowledge

MMLU Pro (Vals)

122 results
  1. #1max of 12292.4%

Leader: this model · 92.4%

knowledge

MMMU Pro (Vals)

81 results
  1. #1max of 8190.6%

Leader: this model · 90.6%

knowledge

Public Benefits Bench v1.1

48 results
  1. #2max of 4874.9%

Leader: Opus 5 · 76.9%

long-context

AA-LCR v1.1

520 results
  1. #4max of 52085.3%
  2. #7medium of 52084.7%
  3. #15high of 52083.7%
  4. #24xhigh of 52083.0%
  5. #39low of 52082.3%

Leader: Kimi K3 · 88.7%

long-context

Arena Document

43 results
  1. #2max of 431513

Leader: Opus 5 · 1516

long-context

MLCR-AA

93 results
  1. #2max of 9371.1%

Leader: Sonnet 5.5 · 75.0%

math

FrontierMath Erdős

7 results
  1. #4max of 70.0%

Leader: GPT-6 Astra · 2.9%

math

FrontierMath Tiers 1–3 (v2)

115 results
  1. #4max of 11590.2%

Leader: GPT-6.1 Sol · 93.7%

math

ProofBench v1.1

50 results
  1. #3max of 50100.0%

Leader: Sonnet 5.5 · 100.0%

other

Arena Vision

116 results
  1. #16max of 1161286

Leader: Fable 5 · 1308

other

T3 Code usage share

85 results
  1. #3max of 859.3%

Leader: Opus 5.5 · 29.1%

reasoning

ARC-AGI-1

215 results
  1. #13max of 21597.5%
  2. #24xhigh of 21596.5%
  3. #28high of 21596.0%
  4. #38medium of 21594.5%
  5. #72low of 21590.0%

Leader: Fable 5 · 98.5%

reasoning

ARC-AGI-2

216 results
  1. #14max of 21690.0%
  2. #15xhigh of 21690.0%
  3. #19high of 21688.8%
  4. #25medium of 21686.3%
  5. #38low of 21678.3%

Leader: GPT-6 Astra · 95.0%

reasoning

Blueprint-Bench 2

28 results
  1. #4max of 2841.9%

Leader: Gemini 4 · 54.4%

reasoning

CritPt

526 results
  1. #9xhigh of 52631.1%
  2. #14high of 52630.3%
  3. #18max of 52629.7%
  4. #21medium of 52629.1%
  5. #30low of 52627.7%

Leader: GPT-5.6 Sol · 32.3%

reasoning

Furniture Assembly

31 results
  1. #5max of 3170.0%

Leader: Opus 5.5 · 83.3%

reasoning

Humanity's Sixth Sense

25 results
  1. #5max of 2540.8%

Leader: GPT-6 Astra · 53.6%

reasoning

LegalBench

134 results
  1. #2max of 13488.5%

Leader: Fable 5 · 88.6%

reasoning

MysteryMechanism

25 results
  1. #4max of 2547.7%

Leader: GPT-6 Astra · 53.2%

reasoning

Roboflow Visual Reasoning

102 results
  1. #30high of 10273.1%
  2. #35low of 10272.0%

Leader: GPT-6 Astra · 91.2%

reasoning

SAGE

80 results
  1. #27max of 8048.5%

Leader: Opus 4.7 · 56.1%

reliability

Tool-call error rate

171 results
  1. #55max of 1710.8%

Leader: GPT-5.4 Pro · 0.0%

reliability

Uptime (7d)

181 results
  1. #74max of 18199.7%

Leader: GPT-5 Pro · 100.0%

safety

CWE-Bench

18 results
  1. #13max of 1850.8%

Leader: Grok 4.7 · 68.3%

safety

CyberBench v1.1

43 results
  1. #19max of 4370.4%

Leader: GPT-6 Sol · 78.0%

safety

SRE Bench

32 results
  1. #9max of 3222.9%

Leader: GPT-6 Astra · 56.9%

speed

Latency (Time to First Token)

355 results
  1. #195low of 3552.9s
  2. #247medium of 3557.4s
  3. #290high of 35521s
  4. #339xhigh of 355116s
  5. #350max of 355290s

Leader: Gemini 2.5 Flash-Lite · 0.3s

speed

Output Speed (tokens/s)

355 results
  1. #231max of 35570.2
  2. #247xhigh of 35563.4
  3. #268high of 35556.9
  4. #274medium of 35555.8
  5. #275low of 35555.8

Leader: Celeris-1 · 1461.1

speed

Provider latency

178 results
  1. #163max of 1784.2s

Leader: Qwen 2.5 7B · 0.2s

speed

Provider throughput

177 results
  1. #122max of 17748.0

Leader: Llama 3.2 3b · 199.5