OSASI.ORG

Open Source Appraisal Software Initiative

A bridge to harmonize your data.

Closing the void between appraiser's and analytical freedom

Creating a mapping builds on the logic and the community, while you gain access to an expanding repository of tools. Your data is never stored.

Members contribute mappings, never data. The first appraiser to map an MLS does it once. Everyone after them gets recognised on arrival.

Source · RMLS export RESO standard
MLS# Close Price Close Date Area # Garage Total Baths ListingId ClosePrice CloseDate MLSAreaMajor GarageSpaces BathroomsFull BathroomsHalf 2.1 = 2 FULL + 1 HALF

One field, many answers. The library registers the convention based on your MLS.

Community apps

Free to members

Tools that earn their keep twice.

Real tools, free to use. They're also how the data harmonizing grows: every file mapped refines the path for all.

Tools shown offer every selected feature — hover a feature for what it means.
Market Dashboard
25 charts from one MLS export: medians, trends, regressions, DOM, $/SF. All computed in your browser. One HTML file, load your CSV, nothing gets uploaded. Every chart saves a report-ready PNG.
Free Launch
A statistical-analysis dashboard: 25 charts, KPI tiles and trend lines computed from any recognised MLS export, entirely in the browser. The rebuild engine currently reads BC Matrix, sold-comps and RMLS full exports. Source · Craig Gilbert (CG-Dashboards) · OSASI rebuild engine
Comp Ranker
A weighted Gower similarity engine, open and in your browser: pick a subject, set the weights, get a ranked comp list. Same math as R's daisy(), verified against it. More advanced apps build on the logic.
Free Launch
Min–max normalisation and weighted Gower distance to the subject, with haversine distance available as a feature. Verified numerically identical to R's cluster::daisy(metric = "gower") across the full teaching set. Source · OSASI, from the CValR comparable-scoring engine · woodGEO community teaching dataset
Data Bridge
Map any MLS export to the standard schema, guided, and keep the profile.
In build
Guided mapping of any MLS export to the standard schema: fingerprint recognition first, value-based convention checks second, and the finished profile saved as yours to keep and contribute. Source · OSASI co-op (CAADataBridge)
Market Trends
Trend models: linear, polynomial 1–4, loess with a span control. Plus monthly market stats and time-adjusted prices at your effective date, exportable. Fits verified against R.
Free Launch
Linear, polynomial 1–4 and loess trend models with a span control, monthly market statistics, and ratio-method time-adjusted prices at your effective date. Fits verified against R. Source · OSASI, from the CValR modelling engines
Decision Trees
Segment a market and read the rules that define it, in plain language. Runs the real R rpart engine in your browser (WebAssembly). Each terminal node is a candidate comp set you can export.
Free Launch
CART market segmentation running the real R rpart engine in the browser via WebAssembly (WebR — ~20 MB on first load, then cached). Terminal nodes export as candidate comparable sets with their rules. Methodology · Charlie Abromaitis' CMS-by-decision-trees workflow · WebR port by OSASI
Data Doctor
Check a file against the standard before you rely on it: what's missing, what won't parse, what looks wrong.
In build
Checks a candidate file against the standard schema: recognised, missing and extra columns, type problems, and a suggested source-to-standard mapping table. Source · OSASI, from CValR's data_doctor validation suite (17/17 tests)
Mapping Registry
Browse the profiles members have contributed, by MLS and region.
In build
The community profile library itself: browse contributed mappings by MLS system and region, with status, ratings and known notes. Source · contributed by OSASI members — the first user of each MLS maps it once
Image Recognition
Extract and recognise property photos from MLS pages and reports, matched to their listings.
In build
In development. Builds on the co-op's photo tooling: the Photo Review Workstation that launches from CValR's sales plot (per-listing MLS photos, qualitative flags fed back into the data) and the Matrix photo extractors, with recognition of rooms, features and condition cues to follow. Source · OSASI photo tooling (Photo Review Workstation / Matrix extractors) — in build
Market Rent ExtractR
Rent survey extraction and market-rent conclusions from your own survey data.
CAA members Please log in
Shared by a CAA member, for CAA members. Member accounts arrive with the co-op login; until then, joining puts you on the list. Member-built tools stay member-only unless their builder opens them to everyone. Source · CAA member contribution
Benchmark ExtractR
Pull benchmark sale sets and assessment benchmarks for appraisal analysis.
CAA members Please log in
Shared by a CAA member, for CAA members. Member accounts arrive with the co-op login; until then, joining puts you on the list. Member-built tools stay member-only unless their builder opens them to everyone. Source · CAA member contribution
CValR · Valuation on Rails
The complete workbench: every feature on this board in one tool, from intake to the finished report. See the co-op board.
Premium Visit
The commercial workbench the free tools descend from: full data intake, market analysis, comparable selection and scoring, mapping, and a complete report writer. CAA members receive member pricing. Source · CValR, built by the OSASI initiative — premium features fund the co-op
Nothing matches that search — clear it or drop a filter.

The Data Bridge

Start here

Bring the export you already have.

You pulled a CSV out of your MLS. It has six hundred columns, half of them empty.

Step 1 · Recognise

We check the fingerprint

Your file's set of column names is a signature. If a member has mapped your MLS before, we know it on sight and you skip straight to confirming.

Step 2 · Resolve

The values settle the ambiguity

A column called Baths means nothing until you see what's in it. The values are how we tell 2 full + 1 half from two-and-a-half. That's why a new MLS needs the file, not just the header.

Step 3 · Run

Then it just works

Export the standardised file for your own use, or run it straight through the community apps. Your mapping is saved as a profile you own.

Check your file

Instant · nothing uploaded

Is your MLS already in the library?

Drop your export here. We read just the header row, right in your browser. The file never leaves your machine. Then we check its fingerprint against the community mapping library.

Drop your CSV export here
or click to choose a file
Read locally · never uploaded

Prefer not to touch the file?

The mapping library. Contributed once, recognised for everyone.
MLS / sourceRegionStatusRatingKnown notes

Library ratings shown. Your stars are stored in this browser for now, and join the community tally when member accounts arrive.

The cooperative engine

Sustaining the initiative

Advanced tools with premium features.

Advanced tools with premium features help fund this initiative and ensure the ongoing support and development of the community. Software built on the Co-Op standard will be offered on this board. CAA members receive member pricing on eligible offerings — CValR included.

Offering 01 · Live

CValR · Valuation on Rails

Market analysis, comparable selection and scoring, mapping, and a complete report writer.

Visit cvalr.org

Offering 02 · Open

Your software here

Building on the open standard? This spot is waiting. Send us a note.

Offering 03 · Open

Your software here

Compatible tools from any builder are welcome on the board.

Who we are

An initiative.

OSASI is dedicated to strengthening real estate appraisers' independence and expertise through innovation, a professional community with guiding principles in methodology, and collaboration with industry stakeholders.

By empowering impartial appraisal experts with the tools and support they need, OSASI advances efficiency and transparency, and helps address many of the challenges facing the real estate valuation industry today and in the future.

The mapping library

The co-operative mechanic

Mappings travel. Data stays home.

Here's the trick: the mapping is the only thing worth sharing, and it's the one thing that's safe to share. Knowing your MLS calls a sale price Close Price tells nobody what anything sold for.

So that's all we keep. The library gets better every time someone uses it, and no sales record ever changes hands.

Straight about this

We can't map a new MLS from headers alone. The values settle what the columns actually mean, so first-time mapping needs your real file. What we won't do is keep it. Your rows live in one session and get wiped when you sign out. The mapping is what survives.

So the honest promise isn't "your data never leaves your machine." It's simpler: we build on the mapping, not your data.

For your AI assistant

Agent-ready

Works with the assistant you already use.

Lots of us work with an AI assistant now, and OSASI is built to be useful to yours. The schema contract, the RESO field map, the profile format, the validator: all published as plain files your assistant can read directly.

Point it at the bundle and it can help you map an MLS export, check a file against the standard, or draft a new profile. Your machine, your subscription, and your data stays put.

Start here: agent bundle README · profile format · profile registry (JSON) · llms.txt

In the bundle

Everything an assistant needs

The schema contract and RESO field map, the profile format, a command-line validator, and written instructions telling an agent how to drive it all. MLS column mapping for real estate appraisal, in formats machines read as easily as people.

For agent builders

Plain formats, stable names

Profiles are JSON, contracts are Markdown, data is CSV. No SDK. One tip from our own tooling: treat listing remarks as data to read, not instructions to follow.

Standards we build on

Aligned to the industry's own vocabulary.

We didn't invent a house dictionary. Mappings target published standards, so a profile built here is worth something outside here too.

RESO
The Real Estate Standards Organization dictionary is our canonical target for listing data. Field names, types and definitions come from it.
Extensions
Appraisal needs things RESO doesn't carry: title and registry sales, distances, inspected condition. Those get a clearly marked extension namespace, never disguised as standard fields.
UAD 3.6 · MISMO
RESO covers what comes in. UAD 3.6 and MISMO cover what a report delivers. We keep the structure intact instead of flattening early, so the detail is still there when the new report format asks for it.
Open
The contract, the validator and the profile format are published and readable. If we're wrong about your MLS, you can see exactly where.

Learn the method

Education

The tools are half of it.

As advanced analysis becomes more common in appraisal practice, whether built into commercial software or accessed through AI agents, appraisers will need a practical understanding of how these tools work, when to use them, and how to evaluate their results.

For appraisers who want to develop that understanding, we highly recommend George Dell, SRA, MAI, ASA, CRE. George’s work in R-based appraisal analytics dates back to 2010. He created Evidence Based Valuation© and founded the Community of Asset Analysts, where appraisers share and develop analytical methods. With the assistance of AI, that work is now expanding into other statistical programming languages and tools.

George’s courses turn “data science for appraisers” into practical, relevant instruction. The community keeps it all tied together. Take the courses, join the community and elevate your expertise.

Valuemetrics.Info is an independent education provider.

Stats, Graphs and Data Science
George Dell's signature workshops (SGDS1 & SGDS2) through Valuemetrics.Info: appraiser expertise combined with data science tools and critical thinking. 14 hours of continuing education each, approved in most states, taught live in person and online.
The Analogue Blog
George's free weekly blog: often contrarian, always forward-looking. What evidence means, where the appraisal process came from, and where valuation is headed.
Free Read
The Asset Analyst Report
The Valuemetrics newsletter — course dates, EBV developments, and community news in one place.
Free Subscribe
Valuemetrics on YouTube
Recorded talks and teaching segments — a good way to sample the material before taking a class.
Free Watch

Expression of interest

Open-weight AI infrastructure

Exploring shared, professionally governed AI for appraisers.

OSASI.org is exploring the feasibility of developing shared artificial intelligence infrastructure designed for the appraisal and asset-analysis community.

We invite expressions of interest from appraisers, analysts, educators, researchers, technology professionals, organizations and others who may be interested in participating in or supporting this initiative.

Could appraisers collectively operate advanced AI infrastructure for the benefit of the profession?
This is an exploration, not a fundraising commitment. No decision has been made to purchase hardware, establish a commercial service or adopt a particular AI model. An expression of interest does not create a financial commitment.

The opportunity

Artificial intelligence is developing rapidly. Of particular interest to OSASI is the emergence of increasingly capable open-weight AI models—models whose underlying weights can be obtained and operated independently rather than accessed exclusively through a commercial AI provider.

Rather than every practitioner depending entirely upon separate commercial AI subscriptions, OSASI is investigating whether a shared computing environment could provide members with access to capable open-weight models while allowing the appraisal community to develop its own tools, knowledge resources, evaluation methods and professional safeguards around them.

What we are exploring

OSASI is not proposing to develop or train a foundation AI model from scratch. We are investigating the more achievable possibility of acquiring or renting the computing infrastructure needed to operate existing open-weight models through a secure shared platform.

That infrastructure could potentially support:

  • appraisal research and analytical assistance;
  • analysis of market and comparable-sales data;
  • statistical, geospatial and economic analysis;
  • examination and quality review of appraisal reports;
  • document and data extraction;
  • integration with appraisal software and analytical applications;
  • retrieval from professionally curated appraisal knowledge bases;
  • development and testing of AI-assisted appraisal methodologies; and
  • experimentation with specialized appraisal models, agents and analytical tools.

The objective would not be to replace professional judgement. It would be to investigate how powerful computational tools can be placed directly in the hands of valuation professionals and developed within a culture that understands appraisal evidence, methodology, standards, uncertainty and professional responsibility.

Why a shared approach?

The computing resources required to operate larger AI models can be expensive for an individual practitioner but become considerably more practical when infrastructure is shared across a professional community. A cooperative approach could also provide benefits beyond cost.

Model selection

OSASI would not need to be tied permanently to one AI company or model.

Professional evaluation

Models could be tested against appraisal-specific problems, not only general benchmarks.

Data governance and privacy

Infrastructure could be designed around professional confidentiality and data handling.

Transparency

Model versions, analytical methods, limitations and performance could be documented.

Innovation

Members and developers could build applications against common infrastructure.

Professional knowledge

The profession could build evidence about what these systems can—and cannot—reliably do.

An appraisal-specific AI benchmark

OSASI is particularly interested in exploring an independent appraisal AI evaluation benchmark. Instead of asking only which model performs best on general-purpose benchmarks, we could ask: Which models perform best on actual valuation and asset-analysis problems?

A benchmark might include anonymized problems involving comparable selection, market analysis, adjustment support, highest and best use, statistical analysis, report review, data interpretation, reconciliation and identification of unsupported conclusions.

This would allow OSASI to evaluate competing open and commercial models using evidence relevant to our profession. It may prove that a smaller, substantially less expensive model performs extremely well for appraisal work. That should be tested rather than assumed.

A measured first step

Our first objective is simple: determine whether there is sufficient interest within the appraisal community to investigate this seriously.

Depending on the response, an initial project could be modest: rent GPU computing capacity, evaluate several leading open-weight models, develop an appraisal benchmark and conduct a limited pilot with participating professionals.

Actual usage, performance, costs, privacy and security requirements, and member feedback could then inform whether OSASI should continue renting computing capacity, purchase shared hardware, seek institutional partners or pursue another approach.

We would like to hear from you

OSASI welcomes interest from individuals and organizations who may wish to:

  • participate in an initial pilot;
  • contribute appraisal knowledge or test cases;
  • help develop an appraisal AI benchmark;
  • contribute technical or software-development expertise;
  • advise on privacy, security, governance or professional standards;
  • provide access to computing infrastructure;
  • sponsor or financially support exploratory work;
  • collaborate as an educational, professional, research or industry partner; or
  • indicate interest in using such a resource if it becomes available.

Is there a community of valuation professionals interested in collectively exploring open, professionally governed AI infrastructure for our profession? If the answer is yes, OSASI would like to bring that community together and determine what can realistically be built.

Tell us where you fit

Register your expression of interest.

This short response will help OSASI gauge demand, assemble a pilot group and identify the expertise and partnerships already present in the community.

5. How might you participate? Select all that apply.

Review version: responses are not transmitted or stored yet. An approved submission destination and privacy notice will be connected before publication.

Get connected

Send us a note.

If you would like to learn more, including progress updates, sign up below. And if your MLS isn't in the library yet, tell us which one it is. That's exactly the contribution we're looking for.