New things you can do in the Smartcat AI chat that were not possible before

You can now get a lot more done just by describing what you need.

The new chat interface changes how you interact with the platform. Previously, every task required navigating to the right screen, filling in forms, selecting settings, and clicking through configuration steps. Now you can describe your goal in plain language, and the system figures out the best way to accomplish it.

Instead of setting things up step by step, you can send files or explain your task — and the system will choose the right workflow and run it for you. Behind the chat, a multi-agent system coordinates the entire process: understanding your request, selecting the workflow, applying your settings, launching the execution, and delivering the result.

What this changes in practice

This makes it much easier to handle things that used to take multiple steps:

Creating multilingual courses from your content. Previously, localizing a SCORM course meant unpacking the package, extracting all translatable elements, managing translations for each component separately, and reassembling everything. Now you can upload a SCORM package and get a localized version back. The system processes the full package — text, images, media, navigation, assessments — and delivers a ready-to-use course.

Localizing websites or courses. Website localization used to involve a multi-step setup process with separate configuration for crawling, extraction, translation, and publishing. Now you can connect your site through the chat, preview translations in context, and launch localization across multiple languages at once.

Translating PDFs and images while keeping formatting. Formatted documents like PDFs and presentations were challenging because translated text needed to fit the original layout. Now you can drop in your file and get a translated version that preserves the original formatting and layout automatically.

Managing content across multiple languages. Working with many languages used to mean managing separate projects or tasks for each pair. Now you can run translation into 10, 15, or 25 languages in parallel. You submit one request and get results for all languages — coordinated and delivered in one place.

Conversational interaction. Instead of navigating forms and screens, you describe what you need in plain language. This is especially useful for ad-hoc tasks, quick translations, and situations where you are not sure which workflow to apply — the system figures it out.

Guided decision-making. You can ask questions about available options, supported formats, or recommended approaches. The system explains what is possible and helps you choose the best path for your specific task.

Automatic use of your existing setups

The system can also use your existing workflows and templates automatically as part of the process. When you describe a task, the system looks at your workspace configuration, finds the matching templates, applies your translation memories and glossaries, and uses your workflow definitions exactly as configured.

You do not need to set anything up again. The system applies your existing configurations automatically — you get the convenience of the conversational interface combined with the precision of your established processes.

What has not changed

All your previous capabilities remain available. The editor, project views, translation memories, glossary management, assignment templates, reporting — everything works as before. The chat is an additional way to access and use these features, not a replacement.

Getting started

Just send your files or describe your task in the chat, and the system will take it from there.