Intento Language Hub for Trados Enterprise

AI translation workflows that meet company requirements in Trados Enterprise

Translators in Trados Enterprise start from drafts that already meet your requirements for terminology, tone of voice, and tags. The Intento Translate step runs multi-step AI workflows that translate, check, and repair each file before post-editing. For each language pair, a Language Setup combines a main and a failover solution that meet your requirements.

Available in the RWS AppStore · Global and EU versions

Long AI workflows

The Intento Translate step waits until LLM-based agentic translation is complete, however many steps it takes.

Meets your terminology and style

Intento AI Agents, built on AI models from 50+ providers, apply company requirements such as terminology, tone of voice, and formatting, and repair what fails.

Three translation modes

Segment-accurate translation keeps translation memory matches, paragraph translation adds context, and subtitles translation keeps timecodes.

650+ languages and dialects

Long-tail languages included, each with a main and a failover solution.

Translators start from drafts that meet company requirements

A review score shows a reviewer which segment misses a requirement, and a person still has to fix it. Intento AI workflows check each translation against company requirements and repair what fails before the file goes back to Trados.

Standard requirements cover general translation quality, approved terminology, tone of voice, formatting, and the tags and placeholders that have to come back intact. Advanced requirements cover company needs beyond that, such as gender-neutral language, terminology that depends on context, profanity filtering, and anonymization.

We receive the full document in the native Trados format, with every inline tag and every piece of protected text. Tags come back in place, and protected text stays unchanged while still giving the translation its context.

Long AI workflows finish inside the Trados project

The standard Machine Translation step is sized for one machine translation pass per request. The Intento Translate step takes its place, after translation memory and before human translation, and waits until LLM-based agentic translation is complete.

In The State of Translation Automation 2026, a three-step workflow of translation, LQA, and repair cut general translation quality errors by more than half, compared with one LLM step with a custom prompt and glossary terms. Such workflows can take over 4 minutes per 1,000 words.

Machine-translated segments arrive with Draft status and Intento as the origin, so reviewers can filter them in the editor.

Example Intento workflows for different content

Each Intento workflow starts from the requirements of one type of content and adds a step only where a simpler workflow misses a requirement. On top, the Trados workflow template with the Intento Translate step; below, the Intento workflow that runs inside it.

Example Intento workflow for technical documentation and product marketing in Trados Enterprise: ten steps inside the Intento Translate step — Technical documentation and product marketing. Ten steps: protected content, translation, LLM post-editing, a code check of tags and placeholders, and a failover LLM from another provider for the segments that fail it.

Example Intento workflow for medical device instructions for use in Trados Enterprise, in segment-accurate translation mode — Medical device instructions for use. Segment-accurate translation keeps translation memory matches. Approved user-interface terms stay in English with the translation in the same tag, and punctuation rules apply where the language needs them.

Example Intento workflow for marketing content in Trados Enterprise, routed by audience to formal or informal Spanish — Marketing content. One workflow per audience: formal “usted” for commercial content and informal “tú” for consumer content, each on a main LLM with a failover LLM from another provider.

Choose the translation mode for each language pair

A Trados admin picks the mode for each language pair in the Intento Translate step, and can keep a workflow template for each mode.

Segment-accurate translationKeeps fuzzy matches above the project threshold and confirmed or approved segments, and translates the remaining segments one by one. For content where the same strings have to be translated the same way.

Paragraph translationTranslates whole paragraphs with their tags, so each sentence gets the context of the sentences around it. For content where fluency matters most.

Subtitles translationKeeps timecodes and one translated cue for every source cue, and takes the display time of each cue into account. For SRT, VTT, and other subtitle files.

The same instruction, translated both ways

A source paragraph of three Trados segments, English → Spanish

	Source	Segment-accurate translationEach segment alone	Paragraph translationThe whole paragraph
1	Connect the power cable to the charging dock.	Conecte el cable de alimentación a la base de carga.	Conecte el cable de alimentación a la base de carga.
2	Place the probe in the cradle.	Coloque la sonda en la cuna.	Coloque la sonda en el soporte.
3	Leave it there until the status light turns green.	Déjelo allí hasta que la luz de estado se ponga verde.	Déjela allí hasta que la luz de estado se ponga verde.
	Why	Without the charging dock in view, “cradle” reads as a baby cradle, cuna. “It” has nothing to refer to, so the pronoun defaults to the masculine.	The charging dock makes the cradle a holder for the device, soporte. “It” is la sonda, a feminine noun, so Déjela.

Solutions that meet your requirements, with any AI model, including your own

For each language pair, a Language Setup combines a main solution with a failover solution from another provider. A solution can use one AI model or several, from the providers we work with or hosted by your company. When new models make a better solution possible, we update the Language Setup, and the Trados workflow template stays the same.

One connection between your company AI gateway and Intento makes company-hosted models available to Trados projects and to the other systems connected to Intento, without a custom TMS integration.

Language Setups also cover long-tail languages. With 650+ languages and dialects, a Trados project can add a less widely served language with its own Language Setup.

The same requirements apply in Trados and across the company

Trados handles the content that goes through localization. The same terminology and tone of voice rules apply in the other systems connected to Intento, such as ServiceNow for employee support and Genesys Cloud for customer chat and email. Our Trados workflows add tag handling on top of the shared rules.

If you own localization, your approved terminology also covers employee and support translation.

Language Hub for ServiceNow →
Language Hub for Genesys Cloud →

Intento Language Engineers can build the solutions for you

If you need help, Intento Language Engineers identify your content groups and the requirements for each, then build and evaluate multi-agent AI solutions for each language pair on your content. They keep the Language Setups current as models change and as translators post-edit in Trados.

The app runs inside your Trados Enterprise workflows

A Trados admin adds the Intento Translate step to a workflow template, in place of the standard Machine Translation step.
Intento receives the full document in the native Trados format, with inline tags and protected text.
Machine-translated segments arrive as Draft, with Intento as the origin, for post-editing and review in Trados.
The Global and EU versions of the app connect to the two Intento data centers.
A Trados admin sets the mode for each language pair, and teams can keep a workflow template for each mode.
Intento is ISO/IEC 27001 certified and GDPR compliant.

Trados is a trademark of RWS. Product names appear on this page only to identify the products and don’t imply endorsement.