The language registry

The human languages NADA covers.

NADA is one shared standard for showing code in a human language. This is the public record of that standard: every language we map, and how far its coverage has come. 130+ languages are usable today across 135 mapped languages, built from 4,182,470 translated terms.

A standard earns trust by being legible. Figures as of June 2026. Coverage grows continuously — this page is the snapshot, not the ledger.

Reading the register

Three states. No grey areas.

Every language sits in one of three states. We publish the state, not a number — a single percentage is easy to misread and hard to keep honest across 135 languages. Exact per-language coverage is available on request [TO CONFIRM: per-language coverage figures / where they are published].

Complete Covers the words used in real code. Usable end to end; refined as vocabulary grows. ●

In progress Past the early threshold, still filling gaps. Many of these already cover more than 70 percent. ○

Planned Mapped and queued. Awaiting a contributor or council review to bring online. —

The register

Coverage, language by language.

A selection from the 135 mapped languages, grouped by script family. Names appear in English and in the language's own script. The full machine-readable register is part of the open language data [TO CONFIRM: link to the published register / dataset].

European

Spanish Español Complete ●

French Français Complete ●

German Deutsch Complete ●

Portuguese Português Complete ●

Italian Italiano Complete ●

Russian Русский Complete ●

Polish Polski Complete ●

Ukrainian Українська In progress ○

Greek Ελληνικά In progress ○

South & Southeast Asian

Hindi हिन्दी Complete ●

Bengali বাংলা Complete ●

Indonesian Bahasa Indonesia Complete ●

Tamil தமிழ் In progress ○

Telugu తెలుగు In progress ○

Thai ไทย In progress ○

Vietnamese Tiếng Việt Complete ●

Marathi मराठी Planned —

East Asian

Chinese (Simplified) 简体中文 Complete ●

Japanese 日本語 Complete ●

Korean 한국어 Complete ●

Chinese (Traditional) 繁體中文 In progress ○

Middle Eastern & African

Arabic العربية Complete ●

Persian فارسی In progress ○

Turkish Türkçe Complete ●

Hebrew עברית In progress ○

Swahili Kiswahili In progress ○

Hausa Harshen Hausa Planned —

Amharic አማርኛ Planned —

Yoruba Èdè Yorùbá Planned —

This excerpt is illustrative, not exhaustive. 130+ languages are usable today; 79 already pass 70 percent coverage. Proprietary or restricted languages are intentionally not listed.

How a language is added

From a single word to a complete language.

Every language starts from English and is built up term by term, by people who read it natively. Each contribution is reviewed before it enters the standard, so coverage grows without drifting. No single person — and no company — decides what a word should be.

01 A term is proposed. A native reader maps a software word to the right word in their language — in context, the way it is actually used in code.

02 The AI council reviews it. An independent panel of models checks the proposal for sense, consistency, and register, flagging anything that reads wrong to a fluent speaker before a human steward signs off.

03 It enters the language data. Accepted terms join the open, CC-BY dataset and ship to everyone in the next build — counting toward that language's coverage.

04 Coverage moves up a state. As the words used in real code fill in, a language crosses from Planned to In progress to Complete.

A language belongs to the people who speak it. So does its place in the register.

Review thresholds and council composition are documented in the governance memo [TO CONFIRM: link to governance / council documentation].

Add your language

If your language is missing, or only half here, you can change that.

You do not need to be a programmer to contribute — you need to read your language well. Bringing one language to full coverage often lifts several related languages at the same time. The data stays free and open for everyone, forever.

Get involved → Or email hello@nadalang.org

Language data CC-BY · 135 languages mapped · the register is public by design.