At memoQ, we take the security of your sensitive data very seriously. Our memoQ AGT solution currently relies on Microsoft’s Azure OpenAI, a limited-access enterprise service. This service offers the same powerful generative language models as OpenAI while providing the security capabilities of Microsoft Azure. It is important to note that the Azure OpenAI service doesn't interact with any services operated by OpenAI, such as Chat GPT.

This means that your data isn't shared with other customers, OpenAI, or any Microsoft or third-party products or services, and isn't used to train or improve large language models. You can read more about our policy in the memoQ AGT Terms of Service.

memoQ AGT supports the following languages:

Albanian, Arabic, Bosnian, Bulgarian, Chinese, Croatian, Czech, Danish, Dutch, English, Estonian, Finnish, French, German, Greek, Hebrew, Hungarian, Indonesian, Italian, Japanese, Korean, Latvian, Lithuanian, Macedonian, Norwegian, Polish, Portuguese, Romanian, Russian, Serbian, Slovak, Slovenian, Spanish, Swedish, Turkish, Ukrainian, and Vietnamese.

Context are TM, LiveDocs, and TB matches - source and target character count.

memoQ AGT counts only the source and target characters of the context and the source characters of the translated text. It doesn't count the prompt. There's a limit to the context you can use based on your TMs and TBs. Make sure you're not providing too much context information, because this can overwhelm the prompt and result in rejecting the translation request.

We calculate characters by adding:

• the number of plain-text characters in the text you want to translate,

• the plain-text characters from context (like TM, LiveDocs, and TB matches),

• 4 characters for each tag in the text, TM, and LiveDocs (since TB matches only count as plain text).

The source segment to be translated + the source and the target of the TM, LiveDocs, and TB hits.