ABI NexaIntelligence OS
The data & access layer
ΣFundamentals & Filings DataStandardized and as-reported, citedfxExcel Add-inThe cited number in your model{ }Data APIThe whole corpus, programmatic∞MCP ServerThe AI-native connector⛱Research WorkspaceThe whole file, per name
AI-native deliverables
ΣOperating & Valuation ModelsBuilt, sourced, refreshed on the print▤Company Tearsheets & ProfilesThe as-of-date one-pager✎Confidential Information MemorandaThe sell-side document, drafted✎Investment MemorandaThe buy-side write-up, traceable▣Pitch BooksProfiles and precedents, quoted⚟Screening & Idea GenerationScreens across the corpus
Who we serve
BBuy-SideHedge funds & asset managersSSell-Side ResearchEquity researchIBInvestment Banking & Corp DevM&A and coveragePEPrivate Equity & VCDeal teams, diligence to model▣CorporatesStrategy, IR and FP&A{ }Data & Quant TeamsAgents and pipelines
Plans Methodology Resources Sign in Request access
Products
Fundamentals & Filings DataStandardized and as-reported, citedExcel Add-inThe cited number in your modelData APIThe whole corpus, programmaticMCP ServerThe AI-native connectorResearch WorkspaceThe whole file, per name
Solutions
Operating & Valuation ModelsBuilt, sourced, refreshed on the printCompany Tearsheets & ProfilesThe as-of-date one-pagerConfidential Information MemorandaThe sell-side document, draftedInvestment MemorandaThe buy-side write-up, traceablePitch BooksProfiles and precedents, quotedScreening & Idea GenerationScreens across the corpus
Clients
Buy-SideHedge funds & asset managersSell-Side ResearchEquity researchInvestment Banking & Corp DevM&A and coveragePrivate Equity & VCDeal teams, diligence to modelCorporatesStrategy, IR and FP&AData & Quant TeamsAgents and pipelines
PlansMethodologyResources
Sign inRequest access
The platform is live

Start with the source. Build from there.

ABI Nexa turns regulatory filings into a standardized, provenanced corpus of fundamentals across markets, comparable across names and true to what each filer reported, with every number traceable to where it came from. Models, memos, profiles and screens are built on that evidence. No blank sheets, no black-box numbers.

Request access Explore the platform Licensed by organisation
7,500+
pre-built models, and growing
0
wrong on the source-verified sample
29.8M
facts, every one checked
The problem

Finance has data. It has AI. What it lacks is the layer in between.

Buy-side, sell-side and banking teams still spend real hours gathering, reconciling and re-entering data, then rebuilding the same models, comps and memos around it. Terminals give them the raw material. General-purpose AI can accelerate the work, but the provenance chain breaks between source, number and output.

Terminals: the raw material

They hold the data and charge for it. The model, the memo and the screen are still yours to assemble.

AI: speed without proof

General-purpose tools accelerate the work, but a number with no filing behind it is one you re-check before your name goes on it.

The gap: the defensible middle

A workflow that turns source filings into standardized, traceable data, and carries that provenance into every model, memo, screen and profile.

Not just faster research. Defensible research.
Why now

One source. Every surface built on it.

Regulatory filings are public and machine-readable. Whoever turns them into trustworthy, cited, workflow-ready output wins. So we built one provenanced fact layer and put every surface on top of it. The model, the profile, the screen, the memo and the answer all read the same number.

ExtractFilings parsed to facts, each keeping its document reference
StandardizeOne fact model across markets and accounting presentations
VersionRestated and as-first-reported held side by side
ServeThe same fact to every surface, human or machine
Change a vintage, and the model, the profile, the screen, the memo and the answer all move together.
The AI layer

The model never does the math.

On ABI Nexa, AI retrieves and explains. It does not invent the numbers, and it cannot pass off a figure it did not fetch.

The math runs in the engine, not the model

Valuations and calculations run in a deterministic engine that reproduces the same result byte for byte. The model narrates that result and checks it for sense; it never produces the number itself.

Every figure resolves to a source

A guardrail checks that each number in an answer traces back to a retrieved cell in the corpus. If it does not resolve, the answer is refused rather than guessed.

Grounded, with the filing attached

Retrieval is scoped to the companies, periods and documents you asked for, and every claim carries the filing, accession and location it came from.

Tested before it ships

A fixed set of questions with known answers runs on every model and prompt change, so a regression is caught by us, not by you.

Why you can trust it

Everyone asserts accuracy. We publish ours.

We ran the audit and published the method, and the harness reports its own exceptions. That last part is what makes the rest believable. See the methodology.

100%

Zero wrong numbers

479 figures re-checked against the regulator's own structured record. Every one matched, at a 95% confidence interval of 99.2 to 100%.

99.8%

Cited to the right filing

The number is right, and the document we attribute it to is the one the regulator records it in.

100.0%

Identity integrity

Every fact's key recomputes from its own content, with 0 mismatches across 29.8M facts and no sampling.

99.88%

Balance integrity

Assets equal liabilities plus equity. The exceptions are surfaced and listed, not hidden.

Products and solutions

One corpus, many ways out of it.

Reach the data the way that fits your workflow, and let it produce the finished deliverable. Every surface is the same provenanced facts in a different container, so they always agree.

ΣLive

Fundamentals & filings data

The number, and the filing it came from.

Standardized and as-reported fundamentals across markets, every value naming the document behind it.
∞Live

Operating & valuation models

Use one already built, or refresh your own.

Open a driver-based model on any name, or bring your own and update it on the earnings print in minutes. Every historical cell cited.
fxLive

Excel add-in

Your model, refreshed on the print.

Pull cited figures into the model you already use, and recalculate when a new filing lands.
▤Live

Company tearsheets & profiles

A sourced profile, as of any date.

An editable one-pager as of any reporting date, with the financials workbook alongside.
⚟Live

Screening & idea generation

Screen on what was filed, not what was assumed.

A missing figure drops a name from that test rather than scoring it zero.
⛱Live

Research workspace

The whole file, already read.

Statements, segments, KPIs, ownership, filings and the prebuilt model, on one terminal.
{ }Live

Data API

The corpus, programmatic, cited.

The same facts your tools consume, with the filing on every value.
∞Live

MCP server

Ask the corpus, not the internet.

Any LLM or agent answers from the entitled corpus with the filing attached.
Who it's for

The same corpus reads differently depending on your seat.

Buy-side

Initiate in an afternoon, then argue with a model and a memo you did not build. Point-in-time vintages for backtests that do not peek.

Sell-side research

Publish with every figure traceable. Refresh the model on the print, not the weekend after.

Banking & corp dev

Profiles and precedent transactions with terms quoted from the filing, not summarised from a database.

Data & quant teams

An audit-ready data layer for agents. The citation travels with the fact, so your output can prove itself.
Questions

The things people ask first.

How is this different from a data feed?
A feed gives you numbers. This gives you the finished pieces of work, a model, a profile, a screen, a memo, a deal file, all built from those numbers and all from one corpus so they agree with each other. Provenance is the reason to trust it, not the product itself.
What does traceable to its filing actually mean?
Every figure carries the document it came from: the form, the filing date and the regulator's own reference. On a source-verified sample of 479 figures, 100% matched the source and 99.8% matched the exact filing.
Can I keep using my own model?
Yes, that is what the Excel add-in is for. Your structure stays yours and only the facts refresh against the new print. You are never asked to migrate into our template.
Which markets do you cover?
US-listed companies, including foreign issuers that file a 20-F, and India today, on one shared fact model, and growing. That is 7,500+ names with a pre-built model. The coverage page states the live counts, dated to the day they were measured.

See it on your own names.

Tell us the markets and the companies you follow, and we will show you the platform on them.

Request accessSign in