FICO 10T and VantageScore 4.0: Two New Mortgage Credit Scores in 2026

The short version

For the first time in decades, the score your mortgage lender pulls in 2026 may not be the one you have been watching on your phone. Two new models, FICO Score 10T and VantageScore 4.0, are now in the picture for home loans, and they do not read a single-day snapshot of your credit. They read trajectory: roughly the last 24 months of your balances, payments, and utilization, looking at where your numbers have been heading, not just where they sit today. If you are thinking about buying a home, that changes which number matters, and it raises the stakes on whether every line on your three credit reports is actually accurate.

• On April 22, 2026, federal housing agencies (FHFA and HUD) announced what they called a new era of credit score competition: FHA will permit VantageScore 4.0 and FICO 10T for FHA-insured underwriting, and Fannie Mae and Freddie Mac are updating their selling guides to move forward with both models.
• This is not a flip-the-switch, everyone-uses-the-new-scores moment. VantageScore 4.0 has been allowed on conventional loans since July 2025, but real-world use is a limited lender rollout as of mid-2026. Non-participating lenders still use Classic FICO.
• FICO 10T is not yet usable to score a mortgage. The Enterprises expect to publish historical 10T data in summer 2026; actual lender use comes later, with no firm go-live date set.
• Tri-merge still rules. All three bureaus (Equifax, Experian, TransUnion) are still required. The earlier proposal to allow two-bureau bi-merge reporting was reversed.
• Both new models read trended data, about 24 months of behavior. Rising balances can score you lower than someone at the same utilization who is paying down. These models can move some borrowers up and some down.
• Because the models read two years of detail, the factual accuracy of every line on your reports matters more than ever. You can pull all three reports free at AnnualCreditReport.com and dispute factual errors under the FCRA.

What actually changed with FICO 10T and VantageScore 4.0 in 2026?

The headline is real, but the fine print matters, so let us be precise.

On April 22, 2026, FHFA and HUD jointly announced that the mortgage market is moving toward credit-score competition. Concretely: FHA will permit VantageScore 4.0 and FICO 10T as eligible models for FHA-insured underwriting, and Fannie Mae and Freddie Mac are updating their selling guides to move forward with both. FHFA titled the announcement Homebuying Advances Into New Era of Credit Score Competition.

Here is what that does not mean:

So the real choice that is emerging is about the scoring model. An approved lender may use Classic FICO or VantageScore 4.0 on a given loan (not both on the same loan initially), not about how many bureaus get pulled. Two different things; only the model choice is live, and only for limited-rollout lenders on the VantageScore side.

• It does not mean every lender now uses these scores. VantageScore 4.0 was allowed for conventional (Fannie/Freddie) loans back in July 2025, but operational use is a limited rollout. Only participating lenders can deliver VantageScore-4.0-scored loans right now; everyone else continues using Classic FICO via tri-merge until it is made broadly available.
• It does not mean FICO 10T is live. The Enterprises plan to publish historical 10T scores in summer 2026 (covering loans acquired roughly April 2013 to September 2025). That is a data publication, not a date you can be scored with 10T on a new mortgage. Broader adoption comes at a later date.
• It does not change tri-merge. You may have read that mortgages were moving to a two-bureau bi-merge model. That proposal was reversed. As of mid-2026, all three bureaus are still required.

What exactly does trended data measure?

Classic FICO is essentially a snapshot. It looks at where your credit stands roughly right now: your current balances, your current utilization, whether you are currently late.

FICO 10T and VantageScore 4.0 add trended data, a time series of your behavior over (typically) the previous 24 months. Instead of what is your balance today, they read:

In other words, they read the direction you are traveling, not just your current location. Experian notes some analyses reference shorter windows (3, 6, or 12 months), but 24 months is the headline lookback both models incorporate, and it applies to revolving and installment tradeline behavior.

• How your balances have moved month to month
• How much you actually paid each cycle (for example, whether you paid only the minimum or well above it)
• How your credit limits and utilization have shifted over time

Can rising balances lower your mortgage score compared to paying down?

This is the part that surprises people, so it is worth stating plainly: trended models can score some borrowers lower than the old snapshot would.

Picture two people with the identical utilization rate today, say both are using 40% of their available credit. Under Classic FICO's snapshot, they look the same. Under a trended model, they may not:

As Experian puts it, even with the same utilization today, the person paying down might present less risk. The trend itself is a risk input.

Mortgage-industry educational material has floated estimates that some rising-balance borrowers could land 10 to 30 points lower than their Classic FICO score, but treat that magnitude as illustrative, not authoritative; it comes from lender-education content, not from FICO, VantageScore, or a regulator. What is solidly established is the direction: trended models can lower some borrowers' scores, not only raise them. A useful corollary: because two years of history persists in the window, paying everything off at once does not instantly erase the prior trend. The picture improves over time, not overnight.

• Person A has been paying balances down over the past two years. Their trajectory reads as an improving financial position.
• Person B has been steadily running balances up to arrive at that same 40%. Their climbing trend can read as possible financial stress.

Can alternative data like rent payments help borrowers with thin credit files?

There is a genuinely inclusive side to this. VantageScore 4.0 is the first tri-bureau model to combine trended data with alternative data; it can factor in rent, utility, and telecom payment history. That is why it can sometimes score thin-file or credit-invisible consumers (young adults, new immigrants, post-bankruptcy rebuilders) that snapshot models simply cannot.

VantageScore says its models can score on the order of 30 to 35 million more U.S. adults than conventional models (the exact figure varies by study and year, so read it as a range, not a constant; these are VantageScore's own vendor figures). FICO 10T's alternative-data reach is narrower: it uses traditional plus trended credit data and reflects rental data only to the extent it is already reported into the bureau file (FICO's broader telecom/utility scoring lives in a separate product, FICO Score XD, not in 10T).

One critical catch: alternative data is not pulled in automatically. A model only benefits from your on-time rent if that rent has been reported to the bureaus and appears in your file. Most landlords report only delinquencies and collections, not positive history, and by recent VantageScore-cited counts, only about 13% of consumers have any rental tradeline on their file (up from roughly 11% in 2024; figures vary by source, so treat this as an approximate, source-dependent estimate). So most renters see no rent benefit unless they actively enroll in a rent-reporting service. And it cuts both ways: a reported late rent payment can hurt.

What does the estimate that 10% more borrowers could qualify actually mean?

You may see a striking projection: that the new approach could help roughly 10% more borrowers qualify. Here is the honest framing.

That figure is an Equifax estimate. The exact wording is that VantageScore 4.0 could help qualify 10 percent more borrowers by adding enriched data like rental history and trended credit. Three things to hold onto:

And the counterweight is mandatory: the same models will score some borrowers lower. Consumer advocates have said so directly. Chi Chi Wu of the National Consumer Law Center cautioned that the change makes very marginal improvements that might help a small fraction of folks, that what is holding back a lot of folks from home ownership is not the credit score, it is the price of buying a house, and warned that negative rent data really harms a renter's ability to get new housing. A credit-repair expert quoted by Fox Business added that reported rent cuts both ways and that newly-qualifying borrowers are probably still coming in at a pretty subpar credit score.

The takeaway is not this is good or this is bad. It is that the new models read more of your history, in more directions, up and down.

• It is a vendor projection about its own product (Equifax), not an independent third party and not a regulator.
• The operative word is could. It is not a guarantee that you will qualify; eligibility still depends on lender overlays, debt-to-income, income, and more.
• It is paired by the same source with other vendor claims (a 15% higher success rate at predicting defaulters, potential $1B in savings), likewise projections, not promises.

Why report accuracy now matters even more

Here is where it gets practical. If a model reads a single snapshot, one stray error affects one moment. If a model reads 24 months of trended behavior across all three bureaus, an inaccuracy (a balance reported wrong, a payment marked late that was not, an account that is not yours, a paid debt still showing a balance) can distort the entire trajectory the model is reading.

A trend built on a factual error is a wrong trend. That is why, under these models, verifying that every line on your reports is factually accurate is no longer housekeeping; it is directly upstream of the picture a mortgage lender sees.

How to pull all three reports for free

The only federally authorized free source is AnnualCreditReport.com. You are entitled to your reports from all three bureaus (Equifax, Experian, TransUnion), free, weekly. Because trended models and tri-merge both span all three, pull all three; an error can sit on one bureau and not the others.

What counts as a disputable factual error

You dispute facts, not your score and not the model. Genuinely disputable items include:

What you cannot dispute away is accurate negative information. A correctly reported late payment or collection is not an error just because it hurts; accurate items stay.

• An account that is not yours (or signs of identity mix-ups)
• A payment reported late that you actually paid on time
• A balance, credit limit, or status reported incorrectly
• A debt you have paid still showing an outstanding balance
• Duplicate tradelines, or an account showing the wrong open/close date

Your two FCRA channels

The Fair Credit Reporting Act gives you two routes for a factual error:

The hard limit to understand: the FCRA guarantees a process, a reasonable reinvestigation, not an outcome. It does not guarantee deletion. If the information is verified as accurate, it stays. The right is to have factual errors investigated and corrected, not to have unfavorable-but-true items removed.

• Section 611, dispute with the credit bureau. When you dispute, the bureau must conduct a reasonable reinvestigation (generally within about 30 days) and forward your dispute to the company that furnished the data.
• Section 623, dispute with the furnisher (the lender, bank, or collector that reported the item). They have their own obligation to investigate disputes received through the bureau.

Where Athena Access fits

The confusing part of all this is not your rights; it is reading your own reports. Tradelines are written in dense, abbreviated bureau shorthand, and a real factual error often hides in a single cryptic line.

That is the narrow thing Athena Access does: paste one line from your credit report, and our free micro-audit helps you decode what it actually says and flag whether it looks like a disputable factual error. We do not promise a higher score, we do not promise you will qualify for anything, and we do not touch the scoring models. We help you understand a line you are staring at and point you toward your FCRA dispute rights if something looks factually wrong. Under models that now read two years of your history, getting those lines right is simply worth more than it used to be.

The bottom line

Two new scoring models are entering the mortgage market on a phased, lender-by-lender basis: VantageScore 4.0 in a limited rollout, FICO 10T not yet live for scoring. Both read roughly 24 months of trended behavior, which means they can move some borrowers up and some borrowers down. Vendor projections about more borrowers qualifying are estimates, not promises. What you actually control is whether your three reports are factually accurate, and with models reading two years of history, that accuracy matters more than ever. Pull all three free, read every line, and dispute genuine factual errors using your FCRA rights.

This article is general financial-education content, not legal, financial, or credit-repair advice. Credit scoring models, mortgage eligibility rules, and rollout timelines change and vary by lender; verify current details with the relevant agency, bureau, or a qualified professional before acting. Athena Access does not guarantee any change to your credit score or your eligibility for any loan.