Seven sheets and 13 slides from the cheapest setting: the two-step guide and the exact prompts I use for Excel, PowerPoint, and Word.

Fable 5.1 has changed how I think people should use frontier models for knowledge work. I asked it to research GoPro’s pending merger with Starman Optical, build a post-close discounted cash flow model in Excel, and turn the analysis into a PowerPoint. On Low, from one short prompt, it produced a seven-sheet workbook with working formulas, scenarios that actually moved the valuation, and a 13-slide deck I could use.

Start it on Low and let it build the workbook, the deck, or the draft. Then open what it made, and move up to High or Extra only for the part that needs judgment. The reason this works is that the Low run is good enough that you can see what’s wrong with it, which means you spend the expensive tokens after you understand the problem instead of before.

I opened the files. There was no Sources sheet and no Checks sheet, so I could follow the analysis but had to work to trace where the assumptions came from. That’s when the Extra run had something to do, and Sol turned out to be better at a different part of the job.

Why does the pattern matter more for knowledge work than it does for code? Because the hard part of knowledge work is deciding whether the work is right. Code has unusually good feedback: the program runs, the test passes, or an error tells the model what failed. PowerPoint, Excel, and Word don’t tell you when the argument is weak, and a spreadsheet will calculate a bad assumption perfectly. The only way to get to clarity is to work through the file itself. Building the spreadsheet changes what I understand about the business, building the deck changes what I think the story is, and writing forces me to decide what I actually believe. You discover the story as you go.

That’s why Fable 5.1 matters. Low is good enough to make each of those iterations useful, and cheap enough that you keep going. It can build the workbook, take my reaction, and revise, without me paying for the most expensive setting on every turn.

This article is the practical companion to my Fable 5.1 review video. The video covers what the model can do. Here I want to get into how I’d actually use it.

Here’s what’s inside:

• What Fable 5.1 did on Low. The seven-sheet workbook, the 13-slide deck, the two sheets it left out, and the one thing Sol still did better.

• Why start Fable 5.1 on low effort. What the five effort settings actually change, what they cost, and the one warning about running research on Low.

• The Excel, PowerPoint, and Word prompts I would use. Including the line that stops Fable from rebuilding a whole workbook when you asked for a small change.

• What I learned about Fable 5.1 as a writer. A 100-word test across three models, and how to separate the edits you want from the ones that flatten your voice.

• What I would tell Fable 5.1 on long jobs. The instructions I add before a run that has to finish on its own, from a 37-second architectural film to a short film about Microduck.

• The exact prompts, in one place. The companion guide, so you can run the workflow without copying anything out of the article.

Let’s start with what Low actually produced, and what it left out.