Model ML, founded by brothers Arnie and Chaz Englander, automates the demanding "last mile" of financial analysis using GPT-5.6 Sol. Their agents carry workflows from initial request through research, modeling, and finished PowerPoint decks or Excel workbooks with traceable sources.

In Model ML's Composite benchmark, GPT-5.6 Sol completed PowerPoint workflows in 100% of test cases versus 76% for Opus 5, and cleared the professional-readiness gate in 43.3% of cases versus 26.7%. It used 36% fewer tokens per workbook than Opus 5 in Excel workflows and 21% fewer than Fable 5 for PowerPoint.

At a global asset manager, a bespoke tearsheet that took an analyst an hour now takes five minutes. Agents processed virtual data rooms with over 100,000 rows and hundreds of files in one pass. The platform is "surface-agnostic," working across email, app, and Microsoft Office plug-ins.

Model ML is moving toward browser-based interactive outputs that stay connected to source material. As Chaz Englander notes, "PowerPoint, Excel, and Word were designed for a world where creating knowledge work was manual. AI has changed that assumption."