Black Friday 23 Nov: Insider US$59 a yearRemind me
Back Back
Models6/day by emailPrimary

In the DeepFinance benchmark of 21 end-to-end private equity and investment banking tasks that take a human professional an average of 27.1 hours, the best configuration, GPT-6 Astra with max reasoning, satisfies 57.6% of rubric criteria, and the top four configurations are not statistically distinguishable

The DeepFinance benchmark, published on arXiv on 7 Nov 2026, measured the percentage of rubric criteria satisfied by 22 model configurations on 21 end-to-end private equity and investment banking tasks. The best configuration, GPT-6 Astra with max reasoning, satisfied 57.6% of rubric criteria, and the top four configurations were not statistically distinguishable.

Verified 2026-11-14

Track with Insider

Why it matters

Models can find relevant data and calculate accurately yet still fail to use that information correctly in complex decisions.

The AI boom,with receipts.

6 sourced facts every weekday, plus the Weekly Brief, AI company deep dives and so much more. Free.

Free · No card · Unsubscribe anytime · See today's issue →

Join now and you start with 100 Receipts on The Receipts Ledger — your first step toward a free year of Insider or Pro. How it works →