XLIFF translation built for localization workflows

Learn about DeepL's quality scores, including a 3-year 345% ROI per a Forrester TEI Study

XLIFF's structure means understanding a few behaviors up front. That way, your translated file comes back in a way that fits your workflow:

• Existing "target" values are overwritten: If your file already contains translations, remove or split out those units first if you need to preserve them
• Per-file language attributes are ignored: The source language/target language attributes on "file" don't control the translation. DeepL uses the API's source_lang and target_lang for the entire upload.
• The state attribute isn't updated automatically: If a segment is marked state="needs-translation", that value is preserved as-is after translation. Update it yourself downstream if your pipeline depends on it.
• All uploaded content counts toward billed characters, even content that isn't translated: Segments marked translate="no", or "target" content that isn't touched, still count toward your character total
• An unsupported trgLang on the root "xliff" element is rejected, even if your API target_lang is valid: Check this attribute if you get an "Invalid target language" error

Head to the web translator, or integrate directly via the API, then follow these steps:

• Log in to your DeepL Pro or DeepL API account
• Head to the the "Translate files" tab
• Upload or drag-and-drop your XLIFF file
• Set your source_lang, or let DeepL detect it per segment
• Select your target language
• Click Translate
• Download your translated file once complete

XLIFF exists because localization workflows need more structure than plain text. DeepL translates it without cracking that structure.

• Localization engineers: Integrate XLIFF translation into an existing pipeline via the API, with segment-level handling that respects your tooling's structure
• Language service providers (LSPs): Handle client overflow or high-volume batches without manual CAT-tool re-entry for every file
• TMS and CAT-tool users: Translate exported XLIFF segments directly, then re-import into your existing translation memory or QA workflow
• Enterprise localization teams: Standardize XLIFF translation across product, marketing, and support content without a per-team custom process

XLIFF workflows rarely stop at one file type or one language pair. DeepL Pro is built to handle the batch.

• Translate PDF, PPT, Word, Excel, HTML, SRT, TXT, XLIFF, JSON, XML, DITA, MIF and IDML files
• Translate several files and into several target languages at once
• Enable a glossary so brand and technical terminology stay consistent across every segment
• Choose between formal or informal tones
• Benefit from maximum data security

Need to translate PDFs, images, and other file types? DeepL’s Language AI adapts visual content to match, so your translations stay consistent and correctly formatted no matter which file you're working with.