Overview

This page outlines the standard General Translation uses to represent locales and languages, as well as the list of currently supported locales.

---

Locale strings

General Translation uses a variant of the BCP 47 language tag standard to represent locales and languages. BCP 47 language tags are the Internet Best Current Practice (BCP) for identifying languages in both spoken and written forms. These tags provide a consistent way to specify languages, allowing applications to adapt content, format, and behaviour based on the user’s locale.

Language tags comprise one or more subtags separated by the "-" character. The subtags include the following components:

• Language subtag: Represents the primary language, e.g., en for English, es for Spanish.
• Region subtag: Specifies a country or region, e.g., US for the United States, FR for France.
• Script subtag (optional): Indicates the writing script, e.g., Latn for Latin script.
• Variant subtag (optional): Identifies a specific variation of a language, e.g., arevela for Eastern Armenian.

When you combine these tags, we refer to this as a locale or locale tag.

Commonly Used Tags

In practice, most language tags consist of two subtags: a language and a region. Here are some common examples:

Extended Tags

Language tags can include additional subtags for greater specificity:

• Example: hy-Latn-IT-arevela
  • hy: Armenian (language)
  • Latn: Latin (script)
  • IT: Italy (region)
  • arevela: Eastern Armenian (variant)

This tag denotes Eastern Armenian written in the Latin script, as used in Italy.

Exceptions to BCP 47 in GT

🚧 This section is currently under construction. 🚧

---

Supported locales

This section lists all locales that are currently supported by General Translation.

A note on low‑resource languages

Our system leverages some of the most advanced LLMs on the market to provide accurate translations; however, these models are not without their limitations. Certain languages may not be supported by the model provider you’ve selected, or by any of the available providers. These are known as “low‑resource languages”.

Low‑resource languages vary between models, so if you specify a preferred model provider in your configuration, you may want to check that provider’s list of supported languages.

Official list

---

Notes

• General Translation uses Locale Tags (locales) to identify languages and regions internally.

Next steps

• See our list of supported locales to find the language tags available in General Translation.
• Refer to the official IETF Language Tag Registry and the BCP 47 language tag standard for more information.