priorsBuilder

Alibaba · released 2026-09-01

Qwen 3.8 Max

Measured on 87 benchmarks across 5 reasoning efforts.

General frontier

70.6

#8 of 136 models · best at minimal

includes estimates: 25% of the mix is measured for this model

Reasoning efforts in your mix

minimal70.6#18 of 313
max47.8#92 of 313
max44.4#113 of 313
xhigh44.3#114 of 313
max43.1#121 of 313

Your mix’s benchmarks

6 of 8 measured

reasoning

Humanity's Last Exam

587 results
  1. #57max of 58743.1%

Leader: Opus 5.5 · 61.4%

knowledge

AA-Omniscience

526 results
  1. #88max of 52612.0

Leader: Opus 5.5 · 46.4

reasoning

BullshitBench v2

196 results
  1. #2minimal of 19695.0%
  2. #131max of 19627.0%

Leader: Opus 4.8 · 95.0%

coding

DeepSWE v1.1

70 results
  1. #37xhigh of 7057.5%

Leader: GPT-6 Astra · 74.1%

math

FrontierMath Tier 4 (v2)

70 results
  1. #30xhigh of 7034.1%

Leader: GPT-6.1 Sol · 100.0%

agentic

Terminal-Bench 4.0

35 results
  1. #25max of 3527.0%

Leader: Opus 5.5 · 64.8%

All 87 benchmarks

show 81 more

agentic

AA-AnalystAgent

44 results
  1. #13max of 4445.0%

Leader: Gemini 3.7 Flash · 60.0%

agentic

AA-Briefcase v1.1

224 results
  1. #16max of 2241617

Leader: Sonnet 5.5 · 1823

agentic

Agent Arena

50 results
  1. #24max of 502.3%

Leader: Opus 5.5 · 15.5%

agentic

APEX-Agents-AA

33 results
  1. #2max of 3342.4%

Leader: Gemini 3.5 Flash · 47.1%

agentic

AutomationBench-AA

219 results
  1. #52max of 21956.2%

Leader: Gemini 4 · 77.5%

agentic

EnterpriseOps-Gym-AA

50 results
  1. #6max of 5047.6%

Leader: Fable 5 · 51.1%

agentic

Harvey's Legal Agent Benchmark

79 results
  1. #18max of 7910.4%

Leader: Muse Spark 1.2 · 25.4%

agentic

ITBench-AA

45 results
  1. #19max of 4540.3%

Leader: GPT-5.6 Sol · 56.2%

agentic

Legal Research Bench

78 results
  1. #13max of 7847.6%

Leader: Muse Spark 1.3 · 55.3%

agentic

SkillsBench

34 results
  1. #27max of 3442.0%

Leader: DeepSeek V4.1 Flash · 69.8%

agentic

Tax Agent Bench

69 results
  1. #19max of 6932.1%

Leader: Fable 5.1 · 49.2%

agentic

Terminal-Bench 4.0 (Vals)

47 results
  1. #15max of 4734.3%

Leader: Opus 5.5 · 65.2%

agentic

Terminal-Bench Science (Vals)

40 results
  1. #12max of 4012.9%

Leader: GPT-6 Astra · 62.9%

agentic

τ³-Banking

206 results
  1. #1max of 20651.3%

Leader: this model · 51.3%

coding

Arena Image-to-WebDev

58 results
  1. #9max of 581638

Leader: Opus 5.5 · 1749

coding

Arena WebDev

122 results
  1. #10max of 1221674

Leader: Opus 5.5 · 1813

coding

Code Migration

77 results
  1. #49max of 7724.0%

Leader: Sonnet 5.5 · 69.8%

coding

FrontierSWE v2

21 results
  1. #18max of 2117.8%

Leader: GPT-6 Astra · 65.5%

coding

IOI

44 results
  1. #12max of 4468.9%

Leader: Gemini 4 · 100.0%

coding

LiveCodeBench (Vals)

127 results
  1. #10max of 12787.9%

Leader: Fable 5.1 · 90.5%

coding

ProgramBench v1 Almost Resolved

62 results
  1. #26max of 6210.5%

Leader: Opus 5.5 · 65.0%

coding

ProgramBench v1 Fully Resolved

62 results
  1. #32max of 620.0%

Leader: Opus 5.5 · 18.5%

coding

ProgramBench v1 Raw Pass Rate

62 results
  1. #47max of 6240.0%

Leader: Opus 5.5 · 87.0%

coding

SciCode

226 results
  1. #76max of 22653.2%

Leader: Opus 5.5 · 66.9%

coding

SWE-bench Verified (Vals)

85 results
  1. #16max of 8585.6%

Leader: Opus 5 · 97.0%

coding

Terminal-Bench 2.1 (Vals)

74 results
  1. #34max of 7467.4%

Leader: Opus 5.5 · 87.6%

coding

Terminal-Bench Science-AA

46 results
  1. #21max of 4611.9%

Leader: GPT-6 Astra · 63.3%

coding

Terminal-Bench v2.1-AA

238 results
  1. #9max of 23888.8%

Leader: Fable 5.1 · 91.4%

coding

Terminal-Bench v4-AA

216 results
  1. #32max of 21638.9%

Leader: Sonnet 5.5 · 63.6%

coding

Vibe Code Bench 1-100

22 results
  1. #19max of 2212.8%

Leader: Opus 5.5 · 30.4%

coding

Vibe Code Bench v1.1

105 results
  1. #43max of 10564.7%

Leader: Sonnet 5.5 · 92.4%

composite

AA Intelligence Index v4.3.2

610 results
  1. #35max of 61045.4

Leader: Opus 5.5 · 57.6

composite

Arena Text

220 results
  1. #21max of 2201483

Leader: Gemini 4 · 1525

composite

Arena Text — Multi-Turn

220 results
  1. #18max of 2201492

Leader: Gemini 4 · 1553

composite

Vals Index

47 results
  1. #34max of 4748.3%

Leader: Gemini 4 · 68.9%

composite

Vals Multimodal Index v1.2

33 results
  1. #12max of 3365.4%

Leader: Fable 5 · 74.2%

composite

Vals RSI Index v1.1

24 results
  1. #13max of 2421.8%

Leader: Opus 5.5 · 37.3%

composite

Vals RSI Index v1.1: Harness Engineering: Judge

24 results
  1. #15max of 2413.5%

Leader: GPT-6 Sol · 22.8%

composite

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

24 results
  1. #11max of 240.0%

Leader: Opus 5.5 · 35.1%

composite

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

24 results
  1. #12max of 2435.3%

Leader: Opus 5.5 · 46.5%

composite

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

24 results
  1. #16max of 2438.3%

Leader: Opus 5.5 · 53.0%

cost

AA Intelligence Index Cost per Task

177 results
  1. #171max of 177$5.41

Leader: GPT-6 Luna · $0.00

cost

Cost to Run AA Intelligence Index

177 results
  1. #163max of 177$4,935

Leader: GPT-6 Luna · $10.63

creative

MineBench

72 results
  1. #50max of 721263

Leader: GPT-6 Astra Pro · 2281

creative

VoxelBench (text)

55 results
  1. #17max of 551787

Leader: GPT-6 Astra · 2732

economics

CorpFin (Vals)

117 results
  1. #27max of 11765.9%

Leader: Opus 5 · 73.2%

economics

Excel Modeling Benchmark

74 results
  1. #34max of 7460.1%

Leader: Fable 5.1 · 76.7%

economics

Finance Agent v2

78 results
  1. #41max of 7850.6%

Leader: Gemini 4 · 65.4%

economics

GDP.pdf

210 results
  1. #44max of 21022.8%

Leader: Muse Wonder Full (Max) · 32.6%

economics

GDPval-AA v2.1

288 results
  1. #16max of 2881667

Leader: Opus 5.5 · 1866

economics

MortgageTax (Vals)

84 results
  1. #46max of 8464.0%

Leader: Opus 5 · 72.1%

economics

TaxEval (Vals)

125 results
  1. #17max of 12575.6%

Leader: Muse Spark 1.2 · 80.4%

knowledge

AA-Omniscience Accuracy

526 results
  1. #166max of 52631.9%

Leader: Fable 5.1 · 67.2%

knowledge

AA-Omniscience Non-hallucination

526 results
  1. #36max of 52671.2%

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

knowledge

GPQA Diamond

539 results
  1. #28max of 53992.8%

Leader: GPT-6 Astra · 96.3%

knowledge

GPQA Diamond (Vals)

122 results
  1. #6max of 12293.7%

Leader: Gemini 3.1 Pro · 95.5%

knowledge

MedCode

94 results
  1. #58max of 9440.7%

Leader: Opus 5 · 63.6%

knowledge

MedScribe

96 results
  1. #31max of 9684.9%

Leader: Opus 5.5 · 91.4%

knowledge

MMLU Pro (Vals)

122 results
  1. #18max of 12288.6%

Leader: Fable 5.1 · 92.4%

knowledge

MMMU Pro (Vals)

81 results
  1. #12max of 8188.0%

Leader: Fable 5.1 · 90.6%

knowledge

MMMU-Pro

265 results
  1. #29max of 26582.8%

Leader: Opus 5.5 · 87.7%

knowledge

Public Benefits Bench v1.1

50 results
  1. #16max of 5067.1%

Leader: Opus 5 · 76.9%

long-context

AA-LCR v1.1

523 results
  1. #69max of 52380.3%

Leader: Kimi K3 · 88.7%

long-context

MLCR-AA

94 results
  1. #30max of 9420.0%

Leader: Sonnet 5.5 · 75.0%

math

FrontierMath Tiers 1–3 (v2)

115 results
  1. #31xhigh of 11565.6%

Leader: GPT-6.1 Sol · 93.7%

math

ProofBench v1.1

52 results
  1. #27max of 5258.0%

Leader: Sonnet 5.5 · 100.0%

other

Arena Vision

116 results
  1. #2max of 1161301

Leader: Fable 5 · 1308

reasoning

CritPt

529 results
  1. #59max of 52920.0%

Leader: Muse Wonder Full (Max) · 32.9%

reasoning

Furniture Assembly

31 results
  1. #31xhigh of 3120.0%

Leader: Opus 5.5 · 83.3%

reasoning

Humanity's Sixth Sense

25 results
  1. #9max of 2533.0%

Leader: GPT-6 Astra · 53.6%

reasoning

LegalBench

135 results
  1. #50max of 13583.6%

Leader: Fable 5 · 88.6%

reasoning

MysteryMechanism

27 results
  1. #17max of 2723.9%

Leader: GPT-6 Astra · 53.2%

reasoning

Roboflow Visual Reasoning

102 results
  1. #15high of 10280.3%
  2. #20low of 10275.9%

Leader: GPT-6 Astra · 91.2%

reasoning

SAGE

80 results
  1. #14max of 8051.3%

Leader: Opus 4.7 · 56.1%

reliability

Tool-call error rate

171 results
  1. #143max of 1715.1%

Leader: GPT-5.4 Pro · 0.0%

reliability

Uptime (7d)

181 results
  1. #84max of 18199.6%

Leader: GPT-5 Pro · 100.0%

safety

CyberBench v1.1

45 results
  1. #45max of 4528.6%

Leader: Wonder · 80.4%

speed

Latency (Time to First Token)

351 results
  1. #199max of 3512.9s

Leader: Gemini 2.5 Flash-Lite · 0.3s

speed

Output Speed (tokens/s)

351 results
  1. #318max of 35142.1

Leader: Celeris-1 · 1522.5

speed

Provider latency

177 results
  1. #124max of 1772.1s

Leader: Qwen 2.5 7B · 0.2s

speed

Provider throughput

176 results
  1. #149max of 17635.0

Leader: Muse Spark 1.1 · 204.3