Blog/Private AI for financial modeling
Use case

Private AI for financial modeling

By Samuel Seidel · Updated September 9, 2026

A model on a dedicated machine can write scenario formulas and catch broken references in a financial model without your revenue, burn rate, fundraising assumptions or unannounced deal terms reaching an AI vendor. Through a consumer chat tool, those numbers sit in the vendor's chat history under whatever terms your plan carries.

Where AI helps with a model

The useful work is the mechanical scaffolding around the numbers, not the numbers. That means writing a formula that references the right cells across a dozen scenario tabs, catching a broken reference after someone inserts a row, drafting conservative, base and aggressive scenarios with a consistent structure for an analyst to fill in, or explaining what an inherited formula computes. A model is also decent at spotting inconsistency: a growth rate in one tab that doesn't match the same line item in another, or a formula that sums the wrong range after a copy-paste.

None of that requires the model to know your numbers are right, only spreadsheet structure and formula logic. In practice, though, the work means pasting real cells, formulas and context into the tool, and that's where the exposure happens.

Why finance is a worse case for a public AI tool

Unreleased financials are one of the more sensitive categories of internal data a company holds: a revenue forecast that leaks before an earnings call, a fundraising model that reveals valuation assumptions before a term sheet is signed, or a cost structure that a competitor could use directly. A cloud AI vendor's terms on training and retention govern what happens to text you send, and they differ by plan and can change. A consumer-tier account typically has weaker guarantees than a negotiated enterprise agreement. For a public company, sending unreleased financials to a third party ahead of a disclosure event also raises questions a compliance team would rather not have to answer.

A model you run yourself takes the AI vendor out of that path. On hardware your organization operates, the numbers stay in-house. On a rented dedicated Spark, they go to your own model on a machine assigned only to you, operated by GPUwerk under a DPA, with no AI vendor retention or training policy applying. If the figures are unreleased, clear the setup under your disclosure policy first.

What this looks like in practice

A general-purpose open-weight model in the 30B-70B parameter range handles formula generation and spreadsheet reasoning well, and it doesn't need specialized fine-tuning for this task since the work is closer to structured text generation than domain expertise. Open WebUI lets a finance team upload a spreadsheet export or paste formulas into a chat interface running against its own model, with no account or API key at an outside AI vendor. For teams that want the assistant to reference a broader set of internal documents (prior board decks, historical actuals, deal terms) alongside the model in progress, the same retrieval-augmented setup used for internal documents generally applies here too.

A single DGX Spark node, with 128 GB of unified memory, comfortably runs a model of this size with enough headroom for a long spreadsheet's worth of context. Finance work is bursty around close, budgeting season or a fundraise, so an on-demand node at $0.79/hour for spark-1x, billed per minute, usually fits better than a reservation. A team running it every day can reserve 1 to 24 months for up to 35% off.

The limits worth knowing

A model doesn't know whether an assumption is realistic; it can only check whether the spreadsheet is internally consistent with the assumptions you gave it. Treat its output as a draft that a human reviews line by line, the same way you'd review a junior analyst's first pass. For what it takes to run private inference for a team beyond this use case, see the private LLM hosting overview.

FAQ

Can a language model actually build a working spreadsheet model?

It can write formulas, generate a scenario structure, and explain what a given line item does, but someone on the finance team still has to check the formulas and own the assumptions. Treat it as a fast way to draft structure and catch formula errors, not as a source of the numbers themselves.

What data actually gets exposed when using a cloud AI tool for this?

Whatever you paste in: unreleased revenue figures, burn rate, headcount plans, cap table assumptions, or the full text of a spreadsheet if you upload it. Chat history and uploaded files sit on the vendor's infrastructure under whatever retention policy applies to your specific plan, and a free or personal-tier account often has weaker guarantees than an enterprise contract.

Does this replace a financial analyst?

No. It speeds up the mechanical parts of building a model, like generating scenario variants or checking formula consistency across tabs, but the judgment about which assumptions are reasonable still comes from the analyst.

Related pages

First top-up: pay $20, get $30 in credit

Keep unreleased numbers out of AI chat histories.

Run a modeling assistant on a dedicated Spark node only your finance team uses. View pricing →

A dedicated DGX Spark in Europe, North America or East Asia, $0.79/hour, billed per minute. $30 of credit is about 37 hours of compute.

Your data stays on a machine assigned only to you. GPUwerk processes it under its published DPA.

Create account See private LLM hosting