— From literary fiction to localization at scale —
Where every decision becomes memory
No cardFirst 5 pages freeYour work never trains AI
Built for work you answer for
Private by default
Your text is never used to train AI, not by us and not by any provider we use, and it stays out of our logs. Delete your account and your text is deleted with it.
How we handle dataSecurity you can check
Hosted in Germany, in Hetzner’s ISO 27001-certified data center. Every sub-processor is listed, and a DPA is one email away.
All sub-processorsGive it context
Glossaries, styleguides, translation memory and per-string notes go with every request. You set them up once per project.
GlossariesYou’re the human in the loop
The AI drafts; you decide. Review every change side by side, keep or revert it, and send a review link when a client needs to look.
Review linksAPI, MCP and webhooks
The full API on every paid plan. Connect Claude or another AI assistant over MCP, or start runs from your own tools.
Developer guideSomething bigger?
Large catalogs, custom workflows, a DPA for procurement. Transept stays self-serve; write to us for the rest.
support@transept.aiWhat it actually does, in six steps
I.One run, every language
Set up the steps once, for example a translation and then a proofread against your glossary. Run them on every language of a document as one job, with a lane for each language and a notification as each one finishes. Anything that needs you waits on the review screen.
II.Your terms and tone, built from your own pages
Transept reads a whole document and proposes a glossary and a styleguide from it. You choose what to keep before anything is saved. Each term has a version for each language, and every new document in the project starts with both.
III.A memory of your decisions and the reasons for them
Every translation you approve is saved with the alternatives you turned down and the comments that decided it. The next run uses that context in every document, yours and your team’s.
IV.Other ways to say any sentence
Click a sentence to see how else it could read, or give a direction like “keep the metaphor”. Your glossary and styleguide still apply, and the rest of the page stays as it was.
V.Give Literess a goal and let her work
“Polish chapter three.” She works through it on her own and rereads the result before calling it done. With Allow All on, her edits apply as she goes, and each one stays in the chat. You can pause or stop her at any point.
VI.Check every change on one screen
Changes from Literess, a proofread or a workflow all wait on one review screen. Keep or revert each one, and when a fix isn’t right, ask for three options. You see the credit cost before anything is generated.
Show how much of it is yours
Working with AI is not the same as letting it write for you. Transept keeps a record of every document: what the AI drafted, what you changed and what you checked. Download it as a PDF when a client, an editor or a publisher asks how the translation was made.
- Counted by word. The words the AI wrote, the words you changed and the paragraphs you wrote yourself, each counted on its own.
- Every step on record. Translations, rewrites, fixes, your edits and your approvals in one list, each line marked as yours or the AI's.
- Your sign-off, dated. Before the PDF downloads, Literess asks whether you read the translation through. Say yes and the report records it.
Two reports on the Free plan, unlimited on every paid plan.
Originality report
Chapter 3 · Ukrainian → English
Reviewed in full by the author, confirmed on Oct 21, 2026
3 languages, 2 AI detectors
We ran AI translations of our own novel, between Ukrainian, English and Chinese, through two popular AI detectors. A detector asks who wrote a text, and a translation keeps its author; Amazon asks how the translation was made. So whatever a detector says, declare an AI translation.
What the detectors saidThree audiences, one workflow
- I.For writers & creatorsYour book, blog or newsletter in another language. The glossary keeps the names, the styleguide keeps your voice, and every line waits for your approval.
- II.For translatorsBatch translation, built-in QA, and a styleguide for each client, so a 50-page job reads as if one translator wrote it.
- III.For localization teamsShared glossaries and styleguides, comment threads on every block, and twelve languages from one source of truth.
Simple pricing
Every plan comes with monthly credits. You spend them only on AI work, at the price each action shows before you click. No sales call and no contract: start free and add a credit pack whenever you need more.
One credit covers one source word in Standard mode. Pro mode uses a more capable model and costs 3 credits per source word. Chinese and Japanese source text is counted per character. Manual edits, exports, comments and translation memory are free: you spend credits only when AI translates, rewrites or checks your text, or reads a document to build a glossary or styleguide. Every paid action shows its price on the button before you click. See the price list.
Free
Always freeFor your first projects
Unlimited users
Translated and proofread on Standard. About five with Pro.
- Pro mode: a more capable model for nuance, 3 credits per source word
- Standard quality: fast, accurate, faithful
- Translation memory: your past translations, reused to keep new ones consistent
- Glossary, styleguide, QA, import / export: all included
- Notion, Drive, TMX / XLIFF: import, export, and bring your own memory
- 68 languages: from Spanish to Tagalog
- Paragraph-by-paragraph translation in every mode
- Literess, your AI literary editor: 10 messages free, then credits
- Originality reports: 2 included
- No card required
Lite
The pocket edition
Unlimited users
- Whole pages and documents in one run
- Unlimited originality reports
- Workflows: chain translation, proofreading, and checks into one run
- API, MCP & webhooks: automate runs, or let AI assistants work in your account
- Teams: create as many as you need and invite anyone for free
- Top up anytime: pack credits never expire
Starter
Most popularFor regular client work
Unlimited users
- 3× Lite’s credits: for weekly work
- Teams: invite anyone for free; add €29 seats to fund a shared credit pool (80,000 credits a month each)
Pro
For higher volumeFor agencies & large catalogs
Unlimited users
- 3× Starter’s credits: headroom for batch work and full catalogs
- Priority support: a direct line and faster replies
- Predictable bill: most months need no extra credit packs
Prices are in euros and exclude VAT. If VAT applies to your purchase, checkout shows it before you pay, and a business with a valid EU VAT ID pays the net price under reverse charge. You can buy credit packs on any plan, including Free: 30,000 credits for €19, 100,000 for €55 or 300,000 for €149. Pack credits never expire. Creating a team and inviting people is free. To share credits, add two or more seats: each seat adds its plan’s monthly credits to the team’s pool, all on one invoice. Cancel anytime and keep your plan until the end of the period you’ve paid for. We give refunds for faults on our side that we can’t fix and for billing mistakes, and EU consumers can withdraw within 14 days and get back the unused part.
What each action costs
Prices are per source word and per target language. Every paid action shows its price on the button before you click. Charges are whole credits, rounded down, and work that fails is never charged.
| Action | Price |
|---|---|
| Translate a paragraph, page or document | 1 credit per source word · 3 on Pro |
| Regenerate or rewrite with an instruction | 1 credit per source word · 3 on Pro |
| Alternatives for a sentence | 1 credit per word of the sentence (3 on Pro), charged only when the options arrive |
| Smart Proofread | 0.2 credits per word it reads (0.5 when it also checks page images), plus 1 credit per word (3 on Pro) for each paragraph it fixes |
| Quality checks (Find) | 0–0.5 credits per word it reads, depending on the check and model; a fix costs 1 credit per word (3 on Pro) |
| AI glossary audit | 1 credit per word it reads inside a workflow; free when you run it from the editor |
| Build a glossary or styleguide from a document | 0.2 credits per source word read |
| Literess rewrites a paragraph | 3 credits per source word, charged when she writes it |
| A message to Literess about your translation (your first 10 on Free are free) | Credits for the work her answer takes; the app shows the range before you send |
| Parallel agents on a long document (Pro plan) | 0.2 credits per word they read, for each update to the shared glossary or styleguide |
| Editing, comments, exports, translation memory, merges, questions to Literess about the app, her own suggestions, the Originality panel and its reports | Free |
| Plan or pack | Credits | Per 1,000 credits |
|---|---|---|
| Lite€9 / month | 25,000 a month | €0.36 |
| Starter€29 / month | 80,000 a month | €0.36 |
| Pro€79 / month | 240,000 a month | €0.33 |
| Small pack€19 | 30,000 | €0.63 |
| Medium pack€55 | 100,000 | €0.55 |
| Large pack€149 | 300,000 | €0.50 |
Nothing could translate what we wrote — so we built what could
Two writers, one problem.
— Vitalii.I wrote 120,000 words of satirical romance fantasy: mages drowning in debt, a multiverse economy that makes Wall Street look ethical, and a love story tangled up in all of it.
— Mariia.I write semi-fictional stories about real people, built around how it felt. Most of the people I wrote about were foreigners. We met in English, spoke in English, lived those moments in English. So I wrote them that way too.
Vitalii VlasiukCo-founder
Mariia IvakhnenkoCo-founder
“But I sometimes wonder what version of them I missed by not writing them in their language.”Mariia
— Vitalii.The same flatness hit me from the other side. Every tool I tried preserved the words but killed the voice. Tone, rhythm, wit, emotional texture: nothing held. It all came back flatter than it went in.
— Both of us.So we did what we know how to do. We read the machine-translation papers. We benchmarked Claude, GPT, Gemini, and a handful of open models on the same 50-page chapter and judged them on accuracy, tone, and whether the satire still landed. None of them held all three; each failed in a different way. So we built the tool we wished had existed when we started.
We still use it every day. This site is localized in Transept: every change to its German and Chinese pages goes through the same workflow you'd use.
FAQ
- Standard is fast and accurate, and the right choice for most content. Pro mode uses a more capable model that keeps more of the nuance, tone, and voice. It costs 3× the credits, which pays off for novels, marketing copy, or anything where the writing itself matters. You can switch per document or per sentence.
- Upload DOCX, EPUB, Markdown, TXT, or PDF files. You can also import directly from Notion or Google Drive, or paste in text. Exports go back to DOCX, Notion, Google Docs, or Markdown, and an EPUB goes back out as the same e-book. A PDF comes in as plain text.
- Yes. You get 5,500 credits a month with no card required, and every mode, Pro included: enough for about 11 pages translated and proofread on Standard, or about 5 with Pro. On Free you translate paragraph by paragraph: click a paragraph, press Translate, read it, move to the next. The core workflow comes with it: glossary, styleguide, QA, comments, and sharing, plus Literess, with your first 10 messages free. Paid plans start at €9 a month and add runs on whole pages and documents, more credits, unlimited teams, and the API, MCP, and webhooks.
- Yes, and it's the core of Transept. Build a glossary for your terms (character names, brand phrases, technical jargon) and a styleguide for tone, style, and rules. Turn them on for a document, and the same terms and tone carry through every translation, from page 1 to page 500.
- Yes. Tell Literess you prefer British spelling, or that you're translating a sci-fi trilogy, and she'll remember it next time. She keeps up to 20 notes per account, and you can review or clear them at any time.
- Yes. Invite teammates, or share a document with specific people to view or edit. Comment on any sentence, resolve threads, and follow changes in the activity feed. No more emailing DOCX files around with "v7_FINAL_final" in the name.
- No. On every plan, your text is never used to train AI models, either by Transept or by any AI provider it uses. Documents are stored in Germany, only the block being translated is sent to the model, and deleting your account deletes your content. The Security & Data page has the details.
- Yes. The Originality panel in the editor counts, by word, how much of the translation you wrote or changed and how much the AI wrote, and lists every edit, fix and approval with who made it. Download it as a PDF originality report: the Free plan includes two, every paid plan has no limit. It is a record of how the text was made, not an AI detector.
- Yes, on every paid plan (Lite, Starter and Pro); the Free plan does not include it. The public API imports documents and string files, runs workflows in every language, and reads the results back. Webhooks tell your systems when a run finishes, and an MCP server lets AI assistants such as Claude work in your account. The developer guide is at transept.ai/developers.
Featured on
