# Build a DCF model from SEC filings with AI

> Ask Investi's analyst for a discounted cash flow model and get one you can change: filed inputs with citations, sliders, scenarios and a sensitivity grid.

Canonical: https://tryinvesti.com/features/dcf-model

Ask Investi's analyst for a discounted cash flow model of any US-listed company. It takes revenue, cash flow, capex, cash, debt and shares from the filings, cites each one, and returns a model with a slider per assumption, bear, base and bull scenarios, and a sensitivity grid.

## Build the model, run the study.

Ask for a discounted cash flow model, a comps table or an event study. The analyst writes the Python, splits a market-wide job across parallel workers, and returns a model you can change: a slider per assumption, scenarios and a sensitivity grid. The inputs are cited; the outputs are estimates.

## Every number links to the filing line.

They cite a news article about the filing; we cite the filing. You see a figure as first reported and as restated, the exact passage or cell, and the calculation. A mismatch check flags a value that doesn't match its source.

## How does the analyst build the model?

It reads the filed figures from the company's XBRL financials, works out the starting assumptions from them, such as five years of revenue growth and free cash flow as a share of revenue, and writes the model in Python. The discount rate and terminal growth are yours to set; nothing in a filing states them.

The model runs in the chat as an interactive figure. Drag an assumption and the value per share, the gap to the last close and the terminal share of value recompute, with the change from the starting case shown next to each.

## What can I ask?

Build a DCF for Costco from its filings and let me change the assumptions. What growth rate does today's price imply? Show the value per share across discount rates of 7 to 10% and terminal growth of 2 to 4%. Compare a bear and a bull case side by side.

## FAQ

**Where do the DCF model's numbers come from?**

The inputs come from the company's own filings: revenue, operating cash flow, capital expenditures, cash, debt and shares outstanding, each with a citation you can hover to see the filing it came from. The discount rate, terminal growth and the projection are assumptions, and the value they produce is an estimate.

**Is the value per share a price target?**

No. It is what the model gives under the assumptions on screen, and it moves when you move them. Investi has no ratings or targets of its own. Investi is a research and information tool, not a licensed investment adviser. Nothing it produces is a personal recommendation.

**Can I build other models the same way?**

Yes. Ask for a comparable companies table, a scenario comparison, payback or loan math, or a study across many companies. The analyst writes the Python, cites the filed inputs, and returns a chart, a table or an interactive model, whichever answers the question.

## Read more

- [Models and studies](https://tryinvesti.com/features#models.md)
- [Introducing the notebook](https://tryinvesti.com/blog/introducing-the-notebook.md)
- [How the analyst cites a filing](https://tryinvesti.com/blog/how-the-analyst-cites-a-filing.md)
- [AI SEC filing analysis](https://tryinvesti.com/features/sec-filing-analysis.md)
