Text and archives · Translation memories and parallel text · Updated 30 Oct 2026

Can a translation agency sell its translation memories to AI companies?

AIData.Broker represents the companies and people that hold translation memories when they sell to AI labs: we value what they hold against public prices, describe it without their name to every buyer that takes it at the same time and help negotiate the terms. Our fee is 12% of the first $250,000 a sale pays, and nothing if no sale closes.

Sentence pairs in two languages from professional translation, with their subject and text type. Translation and multilingual models train on them, and human-translated pairs in most language pairs are scarce.

Value your translation memoriesSelling Books to AI Companies: Prices, Deals and Who Gets Paid
  • No data upload
  • About two minutes
  • No email needed
Prices on record
9

sales that happened, asking prices off the shelf, made to order

Who holds it
Translation and localisation agencies, Companies' in-house localisation teams, Publishers of translated works
Who has paid, offered or listed
Barcelona Supercomputing Center, Pangeanic, ELRA, Linguistic Data Consortium

The short answer · figures from the market's records, last reviewed 30 Oct 2026

Can a translation agency sell its translation memories to AI companies?

Its own, where its client contracts allow reuse, and such text is bought. The Barcelona Supercomputing Center paid EUR 175,500 for a human-quality Catalan-English corpus from Pangeanic, built on a language vendor's own data, and the European Commission paid EUR 1.58 million to have rights-cleared parallel text found and processed for its translation service. Catalogues list parallel corpora for commercial use, such as an English-Vietnamese set at €12,000 from ELRA. The source sentences are the clients' documents, so their contracts decide what the agency may license.

What buyers have paid, offered or asked for translation memories

The whole price board

Every price in the market's records about translation memories, each with its source and the day it was read. Open a row for the source's own words.

9 public prices

  • CA-EN Parallel Corpus: about 15.0 million Catalan-English sentence pairs, human or human-proofread translation, tagged by domain and text type, bought with translation from a language vendor's own database under CC-BY (Barcelona Supercomputing Center, Aina project)

    A realised purchase of existing parallel text plus translation, won by Pangeanic BI Europa 10% under the estimate. The count is the sum of the card's seven text-type counts (583,353 + 6,420,644 + 3,709,344 + 543,010 + 1,379,021 + 1,346,845 + 985,761). The card says the data mixes human translation with machine translation and human proofreading, and was released under CC-BY, so the buyer also bought the right to publish it.

    Text and archivesSales that happenedDec 2022The payer's own page

    TED (EU Official Journal), contract award notice 2022/S 250-731229: Barcelona Supercomputing Center, creation and annotation of datasets to train and evaluate language models (CONSER02022005OP)

    $186K

    per lot (14,967,978 sentence pairs)

    Published as EUR 175,500 (lot 8; estimate EUR 195,000)

    $0.012 per record

  • Amharic-English bilingual news and legal corpus (232,653 Amharic and 291,701 English words)

    ELRA non-member price for a commercial-use (VAR) licence. Per word counts both sides (232,653 + 291,701). A low-resource language pair prices about 100 times the per-word rate of big newswire sets.

    Text and archivesAsking prices off the shelfOct 2026A catalogue or marketplace listing

    ELRA catalogue, Amharic-English bilingual news and legal corpus (ELRA-W0074)

    $4,482

    per corpus, commercial use

    Published as 4000.00 €

    $0.0085 per word

  • Chinese-Vietnamese Parallel Corpus (200,000 sentence pairs, average 15 words per sentence)

    ELRA non-member price for a commercial-use (VAR) licence.

    Text and archivesAsking prices off the shelfOct 2026A catalogue or marketplace listing

    ELRA catalogue, Chinese-Vietnamese Parallel Corpus (ELRA-W0312)

    $2,353

    per corpus, commercial use

    Published as 2100.00 €

    $0.012 per record

  • English-Vietnamese Parallel Corpus (1,000,000 sentence pairs, average 20 words per sentence)

    ELRA non-member price for a commercial-use (VAR) licence. About $0.0003 per word if both sides average 20 words.

    Text and archivesAsking prices off the shelfOct 2026A catalogue or marketplace listing

    ELRA catalogue, English-Vietnamese Parallel Corpus (ELRA-W0311)

    $13.4K

    per corpus, commercial use

    Published as 12000.00 €

    $0.013 per record

  • Chinese English News Magazine Parallel Text (20 million Chinese characters with 9 million words of English translation)

    Non-member fee: research use only, LDC bars non-members from commercial use; a for-profit LDC member pays the lower reduced-licence fee for commercial rights on top of membership. Corpus released 15 June 2005. Per word counts the English side only.

    Text and archivesAsking prices off the shelfOct 2026A catalogue or marketplace listing

    LDC catalog, Chinese English News Magazine Parallel Text (LDC2005T10)

    $2,500

    per corpus (research-only licence)

    $0.00028 per word

  • Chinese News Translation Text Part 1 (474,000 Chinese characters with commissioned English translations, about 285,000 words)

    Non-member fee: research use only, LDC bars non-members from commercial use; a for-profit LDC member pays the lower reduced-licence fee for commercial rights on top of membership. Corpus released 15 March 2005. Human-translated text prices 28 to 54 times the automatically extracted parallel sets (LDC2007T08, LDC2007T09).

    Text and archivesAsking prices off the shelfOct 2026A catalogue or marketplace listing

    LDC catalog, Chinese News Translation Text Part 1 (LDC2005T06)

    $2,000

    per corpus (research-only licence)

    $0.007 per word

  • ISI Arabic-English Automatically Extracted Parallel Text (1,124,609 sentence pairs, about 31M English words)

    Non-member fee: research use only, LDC bars non-members from commercial use; a for-profit LDC member pays the lower reduced-licence fee for commercial rights on top of membership. Corpus released 20 February 2007. Per word counts the English side only.

    Text and archivesAsking prices off the shelfOct 2026A catalogue or marketplace listing

    LDC catalog, ISI Arabic-English Automatically Extracted Parallel Text (LDC2007T08)

    $4,000

    per corpus (research-only licence)

    $0.00013 per word

  • ISI Chinese-English Automatically Extracted Parallel Text (558,567 sentence pairs, about 16M English words)

    Non-member fee: research use only, LDC bars non-members from commercial use; a for-profit LDC member pays the lower reduced-licence fee for commercial rights on top of membership. Corpus released 18 April 2007. Per word counts the English side only.

    Text and archivesAsking prices off the shelfOct 2026A catalogue or marketplace listing

    LDC catalog, ISI Chinese-English Automatically Extracted Parallel Text (LDC2007T09)

    $4,000

    per corpus (research-only licence)

    $0.00025 per word

  • Acquisition of IPR-cleared parallel corpora in all CEF languages, with processing such as anonymisation, delivered to the EU's eTranslation language-resource repository (European Commission DG CNECT, CEF Automated Translation lot 3, SMART 2015/1091)

    The Commission paid a consortium of Tilde, DFKI, the Athena Research Center and ELDA to find, clear the rights to and process parallel text from holders for its machine translation service; the money buys the acquisition work, and the notice states no volume of translation units. Four tenders.

    Text and archivesMade to orderDec 2016The payer's own page

    TED (EU Official Journal), contract award notice 2017/S 001-000038: Tools and resources for CEF automated translation, lot 3

    $1.7M

    per lot (awarded 13 Dec 2016)

    Published as EUR 1,577,911.20 (lot budget EUR 1,600,000)

Raises the price
  • Human translation, reviewed
  • Language pairs that are rare online
  • Tagged by subject and text type
  • Millions of pairs, deduplicated
The rights question

The source sentences are the clients' documents, so what each client contract says about reuse decides which segments the agency may license, and names and confidential details come out first.

Who has bought, offered or listed translation memories

Buyers that take text, and what each publishes

Kind of data
Kind of buyer

7 buyers taking text and archives, largest published figure first

More about text
All text: rates, buyers and rightsForum and community archivesScholarly journal archives

Hold translation memories? A valuation takes about two minutes: it says what buyers pay and whether we would run the sale. Prices last read 30 Oct 2026.

Value your translation memoriesHow a sale works

In this topic

What sells

  1. Start hereWhich data do AI labs want, and who holds it?Each kind of data the market prices, the businesses that hold it, what labs are asking for now and what they turn away.
  2. GuideWhat Are RL Environments? Who Builds, Buys and Prices ThemAn RL environment is a sandbox where an AI agent attempts tasks with tools and a verifier scores each try. Who builds them, who buys them and what they cost.
  3. GuideHow to Sell Your Videos for AI Training: Prices and BuyersAI companies pay to license your videos for training, by the hour or the minute. What each buyer pays, which footage sells, and the rights every frame needs.
  4. GuideSelling audio and speech recordings for AI trainingWhat AI companies pay for call recordings, podcasts, radio, meeting audio and voice data, why archive audio sells cheapest, and the recording-consent law state by state.
  5. GuideSelling photos and image archives for AI trainingWho buys photos for AI training, what stock agencies pay contributors (shares and bonuses, not prices), what image archives sell for, consent and releases agency by agency, and how to prepare a library for a sale.
  6. GuideSelling your company's Slack data for AI training: what the export allows and who buys itYes, a company can license its Slack history to AI buyers. What Slack's export and API terms allow on each plan, whose messages are inside, what buyers pay and the steps.
  7. GuideSelling company documents and shared drives for AI trainingYes, a company can license its documents, wikis and shared drives to AI buyers. Which documents they pay for, what Google Workspace, Microsoft 365, Confluence and Notion let you export, who owns the files, and the steps.
  8. GuideSelling customer support tickets and chats for AI trainingOften yes, after three checks: your customer contracts and privacy notice, your help desk's export terms, and removing personal data. What buyers want in a ticket, what records with tickets fetch, and the steps.
  9. GuideSelling a company's email archive for AI training: exports, consent and priceYes, a company can license its own email to AI buyers. How Google Workspace and Microsoft 365 let you export a whole organisation's mail, the deadlines when a subscription lapses, the law on staff and customer messages, and what buyers pay.
  10. GuideSelling CRM data for AI training: Salesforce and HubSpot records, consent and priceA company can license its CRM history to AI buyers, but CRM records are customer data first. What buyers value in Salesforce and HubSpot, how each lets you export, why the contact list usually stays out, and what CRM history fetches.
  11. GuideSelling meeting recordings and transcripts for AI trainingCan a company license its Zoom, Google Meet or Teams recordings and notetaker transcripts to AI buyers? Recording consent state by state, what Zoom's terms allow, what meeting audio and transcripts fetch, and the steps.
  12. GuideSelling accounting and finance records for AI training: QuickBooks, Xero and NetSuiteA company can license its accounting history to AI buyers, but finance records carry customers', suppliers' and lenders' confidential terms and staff payroll. What buyers value in QuickBooks, Xero and NetSuite books, how each lets you export, what has to come out, and what finance records fetch.
  13. GuideSelling Jira and Linear issue history for AI trainingYes, a company can license its Jira or Linear issue history to AI buyers. It is valued when each issue links to the code that closed it. What each export keeps, what buyers ask for, and what it fetches.
  14. GuideSelling design files and product history for AI training: Figma and beyondYes, a company can license its Figma files and design history to AI buyers. What buyers value, what Figma lets you export, client work, fonts and other assets that are not yours, who owns the designs, and the steps.
  15. GuideSelling IT service desk records for AI training: ServiceNow tickets, runbooks and incidentsOften yes. IT tickets, incidents and runbooks teach agents to resolve real problems, once credentials, hostnames and people are removed. What ServiceNow and Jira Service Management let you export, what buyers ask for, and what the records fetch.
  16. GuideSelling Books to AI Companies: Prices, Deals and Who Gets PaidAI companies license books by the title, mostly from publishers' backlists. What deals paid per book, how much reaches the author, and how to sell a catalogue.
  17. GuideSelling Game Code and Game Data to AI CompaniesAI companies license game studios' code with its history and gameplay recorded with the player's inputs. Who buys, what studios are offered, and what you may not sell.
  18. GuideSelling a company's paper archive for AI trainingYes, a paper archive can be licensed to AI buyers once it is scanned into searchable text. What scanning costs, what text fetches, who owns old records, and the steps.
  19. GuideSelling robot and first-person video data for AI trainingRobotics companies pay for first-person video and teleoperation data, mostly recorded to their spec. What each kind pays, why bulk footage collapsed, consent, and the steps.
  20. GuideSelling sales call recordings and sales emails for AI trainingYes, if each call was lawfully recorded and the prospects come out. Who buys sales calls and sales emails, what they fetch, and the consent and AI-agent rules.
  21. GuideSelling CAD files and engineering drawings for AI trainingYes, if the company owns the designs and export rules allow it. Who buys CAD files and drawings for AI training, what they pay, and what to check first.
  22. GuideCan You Sell AI-Generated or AI-Assisted Data for AI Training?Rarely if a model made it: buyers screen generated data out. Work people made with AI tools still sells when you can show which parts a model wrote.
  23. GuideAI Content Licensing: How Publishers Get Paid, and How MuchAI companies license content to train models or to ground answers. What publishers have been paid, which platforms pay per use, and how to choose a route.
  24. PageWhere data comes fromWho holds each kind of data, how much exists and how it reaches buyers.
  25. RegisterCompanies closingCompanies closing, sold or gone dark, and the records they leave.