• For buyers
  • Docs
Sign inSign up
Blog
Nov 14, 2026Guide

Best Mercor Alternatives for Selling Data to AI Labs

The best Mercor alternatives for selling data to AI labs in 2026: which assets each accepts, how much preparation it handles, and what deal terms it offers.

By DataVendor team

Mercor is best known for supplying expert contractors to AI labs. It also runs Mercor Data, a program that licenses workplace data from companies. That program suits some sellers well. If you hold a private codebase, want to set your own price, or want to license the same asset to several labs, another route may earn you more.

At DataVendor, we help sellers bring code and data to AI labs. This guide compares the main alternatives to Mercor on three questions. Does the service accept the asset you have? How much of the preparation will it handle? And what deal will you sign, including who sets the price and whether you can sell again? DataVendor wrote this guide. The details about each company come from its own public pages and are current as of November 2026, and where a company does not publish a term, we say so.

Understand what Mercor Data offers

Mercor connects to more than 100 workplace tools, including Slack, Gmail, GitHub, Notion, and Salesforce. Its GitHub connection covers repositories, commits, pull requests, and issues, so Mercor Data handles code as well as workplace records. Its platform extracts and de-identifies the data, and Mercor then licenses the result to buyers. Mercor says no lab or third-party vendor sees your raw data.

Compensation depends on the volume of data, the tools you connect, and the depth of the records. Mercor states that there are no upfront costs, setup fees, or deductions, and that it covers the costs of extraction, anonymization, and transfer. It pays when the anonymized dataset reaches the buyer, typically within two to four weeks of the signed agreement.

Mercor works as a managed vendor. It sets the price for your data, packages it, and handles the sale to labs, so your team does little work. In exchange, Mercor keeps the margin from selling it on, and you are paid once, when Mercor delivers the dataset to a buyer. For a high-quality asset that several labs want, that one payment is often less than the asset could earn through a broker.

Some terms are also missing from the public page. It does not say whether the license is exclusive, which rights you keep after the deal, or which labs buy the data.

Choose a lane before you compare

The options below fall into two lanes. A broker, such as DataVendor, helps you prepare your asset and sells it to several labs on your behalf. Because several labs can license the same asset, a high-quality asset can earn more this way, but an asset that labs do not want will not sell. A managed vendor, such as Mercor, sets the price and handles the sale. It usually pays less for a strong asset, but it does more of the work for you.

So the first question is the quality of your data. If it is high quality, sell it through DataVendor. If you are not sure, get an estimate before you accept Mercor's offer. If the data is unlikely to interest labs directly, a managed vendor that sets the price and handles the sale may be the better choice.

Compare the alternatives

The table summarizes how each option handles the three questions. The sections that follow explain each one in more detail.

OptionHow it worksAssets it acceptsPreparationDeal terms
DataVendorBroker and marketplaceCodebases, including legacy COBOL and Java, with engineering context; documents, logs, and enterprise records; tasks for buyer requestsImports from GitHub, GitLab, and Bitbucket; team-assisted PII removal, secrets cleanup, and packagingYou set the price; license with ownership kept by default; 20% fee at withdrawal; optional checks priced in credits
Turing Project LazarusManaged licensingChat, email, tickets, wikis, CRM, databases, and optional codeTuring de-identifies and packagesNon-exclusive; ownership kept; revenue share after a 90-day sole selling period
Scale AIManaged licensingChat, email, documents, work tools, and data warehousesAnonymized inside your environmentRecurring revenue; fees and exclusivity not published
AppenManaged licensingOperational documents, decision records, and systems of recordExtraction, anonymization, and synthetic rewritingDefined license; ownership kept; paid on acceptance
HandshakeManaged licensingOperational data from any industryDe-identificationLicense; fees and exclusivity not published
TroveoManaged licensingBusiness data and videoAnonymization and clearingOwnership kept; revenue share
SimpleClosure AssetHubShutdown brokerSource code, workspace data, IP, and domainsAssessment, personal-data processing, and packagingLicense or sale; success fee not published
ProximalDirect buyerProduction codebases onlyYou run an evaluation scriptNon-exclusive; ownership kept; offers of $5,000 to $20,000 or more
FileYieldPrivate brokerAny dataset, including codeNoneSale, API access, or license; 8% seller fee

1. DataVendor

DataVendor has relationships with all of the leading AI labs and brokers sales of private code, raw data, and tasks to them. It differs from Mercor in the two places that matter most to a seller with a valuable asset. You set the price, and you can license the same asset to more than one lab, so you are paid each time a lab buys.

DataVendor's platform fee is a share of your earnings, charged when you withdraw them, so its return rises with your price. Its team helps you assess buyer fit, prepare the asset, and list it.

DataVendor fits private codebases with their engineering history best, and legacy enterprise codebases, including COBOL, Java, and mainframe systems, are welcome. It also accepts raw data, such as documents, logs, and enterprise records delivered as ZIP files. Tasks imported through HUD sell through private listings submitted to buyer requests. If a codebase is not ready to upload, you can register it in Inventory for an initial assessment, then attach the code before you list it.

DataVendor imports repositories from GitHub, GitLab, and Bitbucket, with their history, pull requests, and reviews. It also imports project data from Jira and Confluence, though those imports cannot be listed for sale yet. Other records, such as Slack or email exports, upload as raw data, and the team can help package records that need special handling.

DataVendor's team removes personal information and committed secrets from codebases and workplace exports, such as email, Slack, Google Drive, and document exports, with the scope agreed for each asset. You do not need to finish that cleanup yourself before you ask for an assessment.

Before a purchase, buyers see metadata, redacted file trees, and quality evidence, but never the raw files. The codebase estimator gives you an advisory, dated value range before you list. To see current demand, open the Wanted tab after you accept the vendor agreements. It lists open requests from buyers, and you can submit a listing that meets one.

A sale licenses your asset for a commercial use, and you keep ownership by default. DataVendor charges a 20% platform fee on earnings when you withdraw them, and earnings are available as soon as a buyer pays. A buyer who purchases a listing outright stays anonymous, but an offer names the buyer's organization, so you can choose whom to accept.

Prices depend on quality. DataVendor's strict confidentiality policies prohibit us from naming buyers or sellers, or from sharing the figures from specific deals, so the numbers here are directional, current as of November 2026, and not guarantees. About $30,000 per license is the floor for a good codebase, and some deals for a single codebase have exceeded $10 million. A codebase built by several dozen engineers, with thousands of pull requests and over a million lines of mostly human-written code, recently licensed in the low six figures. Poor-quality codebases usually do not sell.

Engineering quality matters as much as size. A codebase with similar history but weaker engineering can sell for about a third as much as one built by top engineers. Labs also pay more for complex products, for five or more years of engineering history, and for code written mostly by people. Our guide to selling source code explains what labs look for.

Strong assets also move quickly. For a complex codebase with a rich engineering record, procurement can take as little as two to four weeks. High-quality codebases from large engineering teams sell to the most lab buyers, so a series of licenses can earn more than one payment for the same asset. Sellers on DataVendor include several publicly traded companies.

DataVendor is the strongest choice when you hold a high-quality codebase or proprietary dataset that labs want. It is not the right place for every asset. Poor-quality data usually does not sell, and lower-quality codebases stay on the marketplace longer. If your data is unlikely to interest labs, a managed vendor that sets the price and handles the sale may serve you better.

2. Turing Project Lazarus

Project Lazarus is the closest managed alternative to Mercor, and it publishes more of its terms. It licenses chat, email, tickets, wikis, CRM records, databases, and source code. Code is optional. Turing says communication, tickets, and documents often matter more.

Turing accepts companies that are operating, winding down, or already shut down. After a mutual NDA, it reviews a real sample and prices from that. It handles de-identification and preparation at no upfront cost and recovers those costs from later sales.

The licenses are non-exclusive, ownership does not change hands, and you owe nothing if nothing sells. You receive a share of the revenue from each sale. Two terms deserve a close look. Turing is the only seller of your record for the first 90 days, after which you can renew or opt out. The page also does not publish the revenue split.

Lazarus suits a company that wants a managed service with non-exclusive terms, including a company that has already closed.

3. Scale AI Data Partnerships

Scale AI's Data Partnerships program licenses de-identified chat, email, documents, work-tool data, and warehouse data. Scale anonymizes the data inside your environment before anything leaves it, which can help a company with strict security rules.

Scale describes the program as an opportunity for recurring revenue. It does not publish its fees, its exclusivity terms, or which rights you keep. It names 40 or more employees as its preferred starting point, so it is aimed at established companies rather than small teams.

4. Appen Enterprise Data

Appen's Enterprise Data program licenses operational documents, decision records, and data from systems of record. Appen extracts and anonymizes the data. Where the structure of a workflow must survive redaction, it rewrites the sensitive parts with synthetic content.

Appen publishes more of its terms than most managed programs. A partnership grants a defined license for a defined use, and it does not transfer ownership. Appen states that there are no upfront costs, setup fees, or deductions, and it pays when the buyer accepts the data rather than when you submit it. It does not publish whether the license is exclusive.

5. Handshake AI Data Partnerships

Handshake buys operational data from businesses in any industry and de-identifies it. Its page advertises earnings of $100,000 to more than $4 million, which its license page describes as indicative rather than an offer. It does not publish fees or exclusivity terms.

If you have a large body of operational data, Handshake can give you a second managed quote to compare with Mercor's.

6. Troveo

Troveo began by licensing video and now licenses business data as well, including chat, email, CRM records, and repositories. It licenses the data on your behalf under terms you approve. Ownership never transfers, and you share in the revenue from each license.

Troveo scrubs names, contact details, credentials, and client identities before licensing. Its fee and exclusivity terms for business data were not available on its public page, so ask for them before you sign.

7. SimpleClosure AssetHub

SimpleClosure's AssetHub sells source code and workspace data from tools such as Slack, Google Drive, email, Notion, and Figma, along with IP and domains. SimpleClosure assesses each asset, processes datasets for personal information, packages them, and matches them with buyers.

A deal can license the data while you keep the underlying assets, or sell the assets outright. SimpleClosure takes a percentage only when an asset sells or is licensed, but it does not publish the rate. Operating companies can use AssetHub. It fits best, however, alongside SimpleClosure's shutdown service, where asset sales become part of closing the company.

8. Proximal

Proximal licenses production codebases directly. You run an evaluation script on your code, and Proximal says it makes an offer within 48 hours. The license is non-exclusive, and you keep full ownership.

Proximal advertises offers of $5,000 to $20,000 or more, depending on the company's stage. That makes it a quick, simple option for a smaller codebase, though not the route to the highest price for a strong one.

9. FileYield

FileYield is a private brokerage that introduces data owners to buyers. It does not handle the data or remove personal information, so your dataset needs to be clean before you start. It charges sellers 8% and buyers 6%, and a deal can be a sale of a copy, API access, or a license.

Consider other options

A few other companies come up in this space, but they serve narrower needs. Commodity data marketplaces such as Datarade and Opendatabay list commercial datasets for a commission. They do not prepare company archives or codebases.

micro1's data partnerships program targets established companies with 30 or more full-time employees. It accepts software and engineering data, such as source code documentation, engineering workflows, and bug tracking, and micro1 de-identifies the data itself. It publishes payment ranges from $100,000 to over $3 million, paid within 30 days of acceptance. Its data license master terms, public as of November 12, 2026, describe a non-exclusive license under which you keep ownership of the underlying data, subject to each order form. Human Native AI joined Cloudflare in January 2026, and it is unclear whether it still accepts new sellers. Surge AI and Snorkel AI supply expert work and datasets to labs, but they do not buy existing company data.

Choose where to start

Start by finding out what your data is worth. Get an estimate for a codebase, or book a call for a quote on other data. The answer tells you which lane fits.

Your situationWhere to go
A high-quality codebase with years of historyDataVendor
A legacy codebase, such as COBOL, Java, or mainframe codeDataVendor, whose team helps prepare it for buyers
High-quality workplace data that labs wantDataVendor, whose team removes PII and lists the data for labs
Data of uncertain quality, or a wish for the vendor to do the workA managed vendor such as Mercor, micro1, Scale AI, or Appen
A wish for someone else to run the sale for a revenue shareTuring Project Lazarus
A company that is closingA DataVendor estimate before accounts close, or SimpleClosure AssetHub for a sale tied to the shutdown
A smaller codebase and a need for a quick offerProximal

Whichever route you take, get the deal terms in writing before you connect systems or share samples. Ask who sets the price, whether the license is exclusive, which rights you keep, and whether you can license the same data again. Our AI data licensing guide lists the terms to agree on.

FAQ

What is Mercor Data?

Mercor Data is Mercor's program for licensing workplace data. It connects to tools such as Slack, Gmail, and GitHub, extracts and de-identifies the data, and licenses it to buyers. Mercor pays by wire when the anonymized dataset reaches the buyer, typically within two to four weeks.

Should I take Mercor's offer or sell through DataVendor?

It depends on the quality of your data. If it is high quality and labs want it, DataVendor can earn you more, because it brokers the sale and can license your data to several labs. Mercor sets one price and handles the sale, so it usually pays less for a strong asset but can be a good choice for data of uncertain value. At DataVendor's 20% fee, you match a $100,000 offer with no deductions at $125,000 in total sales, for example two licenses of $62,500. Get an estimate before you decide.

What is the best Mercor alternative for selling a codebase?

For a high-quality private codebase, DataVendor lets you set the price and license the same codebase to more than one lab. Turing's Project Lazarus is a managed alternative with non-exclusive terms. Proximal makes quick offers on smaller codebases.

Which Mercor alternatives offer non-exclusive licenses?

Turing's Project Lazarus and Proximal publish non-exclusive licenses, and micro1's public master terms describe a non-exclusive license with ownership kept, subject to the order form. On DataVendor, you keep ownership by default, and the rights depend on the agreement behind each listing. Mercor, Scale AI, Appen, and Handshake do not publish their exclusivity terms.

Which options handle data preparation for me?

Mercor, micro1, Turing, Scale AI, Appen, Handshake, Troveo, and SimpleClosure all extract or de-identify data for you. DataVendor's team helps you assess, prepare, and package your asset, including PII removal and secrets cleanup for codebases and workplace exports, and DataVendor imports repositories from GitHub, GitLab, and Bitbucket. FileYield expects you to prepare the data yourself.

Can I use more than one of these services?

Often you can, but check each agreement first. An exclusive license, or a sole selling period such as Turing's first 90 days, can stop you from offering the same data elsewhere.

Compare your offers

Before you accept a set price, find out whether your data could earn more through a broker. Get an estimate for your codebase, or book a call for a quote on other data. If your data is high quality and you want to earn from more than one sale, DataVendor is the right place to sell it.

OlderBest Places to Sell Data to AI Companies in 2026

Estimator tools

  • Codebases
  • Company data

Resources

  • For buyers
  • Docs
  • Blog
  • Contact
  • Careers

Legal

  • Privacy
  • Terms
  • Trust

Social

  • GitHub
  • X
Provided by
  • For buyers
  • Docs
Sign inSign up