Domains: FinanceBusinessMath
Skills: AgenticTool UseKnowledge WorkCodingReasoning

Four-Factor Integrity Breakdown & Primary-Source Evidence

Privacy

Weight 40%
10/10
Access Model: Strictly Private
"AA-AnalystAgent is a privately held benchmark; the question set, reference answers, and source files are not publicly released, to limit contamination risk"

Originality

Weight 25%
7.2/10
Reuse Profile: Low Reuse (<25%)
"80 quantitative questions across 14 business and scientific domains, including environmental reporting, trade and commodity statistics, healthcare expenditure reports, hydrology and weather data, government appropriations, energy cost models, financial models, and project schedules"

Freshness

Weight 20%
10/10
Initial Release: 2026-08Latest Refresh: 2026-08Rolling Updates: No
"August 10, 2026 Announcing AA-AnalystAgent: an agentic benchmark for quantitative analysis on real-world spreadsheets and documents"

Grading

Weight 15%
3.3/10
Integrity: Self-Reported / Live NetworkSandbox: live internet
"The agent is provided a small toolset covering code execution in an isolated Linux sandbox (Python 3.12 with the question's reference files mounted and a pinned set of standard Python data analysis libraries preinstalled), URL fetching, image viewing for vision-capable models, and final-answer submission."

Frontier Model Standings on AA-AnalystAgent

Top 15 active models tracked
Rank Model Official Score Percentile
#1GPT-6.2 Astra max OpenAI62.5%97.2p
#2Gemini 3.7 Flash high Google60.0%93.7p
#3Claude Fable 5.1 max Anthropic57.5%88.3p
#3Claude Sonnet 5.5 max Anthropic57.5%88.3p
#5Claude Opus 5.5 max Anthropic56.2%83.0p
#6Claude Haiku 5.5 max Anthropic53.8%77.7p
#6GPT-6.2 Sol max OpenAI53.8%77.7p
#8Grok 4.7 xhigh SpaceXAI52.5%72.4p
#9GPT-6 Astra max OpenAI51.2%67.0p
#9Inkling 2 Small default Thinking Machines51.2%67.0p
#11GPT-6.1 Sol max OpenAI50.0%61.7p
#12Gemini 4 Argon high Google48.8%58.2p
#13Gemini 3.8 Flash high Google46.2%51.1p
#13Qwen4 Max xhigh Alibaba46.2%51.1p
#16Qwen3.8 Max (0902) xhigh Alibaba45.0%44.0p

Embeddable Audit Badge

View Markdown (.md)

Benchmark maintainers can embed this live SVG badge in GitHub READMEs or documentation:

BenchAudit AA-AnalystAgent Tier A 8.3/10
[![BenchAudit Tier A](https://benchaudit.org/badges/aa-analystagent.svg)](https://benchaudit.org/benchmarks/aa-analystagent/)