Tongue
On-device language detection for text across 84 languages. Identify the language in tens of microseconds, from just a few words.
Demo
Detect language based on 3 words.
"kann ich das haben" is German. "안녕하세요" is Korean. Three words is the hard case, and three words is what a search box gives you.
From three words, Tongue scores 0.933 where a 293MB detector scores 0.887. Tongue is 2MB, and works in two stages: some scripts belong to one language only, so Hangul is Korean and Thai is Thai, settled with no model involved. The rest go to the model.
Tongue needs no tokenizer, vocabulary file, or inference runtime, which is how the whole package fits in 2MB. When the input is too short to call, Tongue reports a tie.
0.933 from a 2MB model.
At 2MB, Tongue scores 0.933 on three-word FLORES-200 input, against 0.887 for a 293MB detector.
Measured on the shipped int8 weights across three public benchmarks, with the other detectors run on identical rows and language subsets. Full tables and failure modes are on the model card.
Use cases
Detect language based on 3 words.
Route text by language
Pick the right translation model, analyzer, or support queue from the first few words a user types, entirely on device.
Keyboards and input
Detect the language mid-sentence to switch autocorrect, suggestions, or layouts, at a cost of tens of microseconds per keystroke.
Search and content pipelines
Tag short strings with a language before indexing or moderation, without a network round-trip or a per-request service.
Ask for more text instead of guessing
"la casa" is Italian and Spanish in equal measure. Tongue reports the tie, so your product waits for another word rather than switching the keyboard by mistake.
Inspiration
Ideas to build with Tongue. Copy a prompt into your coding agent and go.
Route incoming messages to the right language agent.
Use Desert Ant's Tongue to identify the language of an incoming message from a few words, on-device, and route it to the right support queue, model, or translation path. Build it with the Desert Ant SDK. Tongue (Swift, Kotlin, JavaScript / TypeScript). Install and API: https://desertant.com/docs/tongue/. Install: SwiftPM desert-ant-core; Maven ai.desertant:tongue:3.1.0; npm i @desert-ant-labs/tongue. SDK source: https://github.com/Desert-Ant-Labs/desert-ant-core. Machine-readable catalog of every model and SDK: https://desertant.com/llms.txt.
Auto-switch the keyboard and spellcheck to the language being typed.
In a text field, run Desert Ant's Tongue on the first words to detect the language and switch keyboard, spellcheck, and autocorrect to match, in microseconds, on-device. Build it with the Desert Ant SDK. Tongue (Swift, Kotlin, JavaScript / TypeScript). Install and API: https://desertant.com/docs/tongue/. Install: SwiftPM desert-ant-core; Maven ai.desertant:tongue:3.1.0; npm i @desert-ant-labs/tongue. SDK source: https://github.com/Desert-Ant-Labs/desert-ant-core. Machine-readable catalog of every model and SDK: https://desertant.com/llms.txt.
Tag user posts by language for search and moderation.
Use Desert Ant's Tongue to label each user post with its language on-device, so search, ranking, and moderation can branch by language without a server call. Build it with the Desert Ant SDK. Tongue (Swift, Kotlin, JavaScript / TypeScript). Install and API: https://desertant.com/docs/tongue/. Install: SwiftPM desert-ant-core; Maven ai.desertant:tongue:3.1.0; npm i @desert-ant-labs/tongue. SDK source: https://github.com/Desert-Ant-Labs/desert-ant-core. Machine-readable catalog of every model and SDK: https://desertant.com/llms.txt.
Detect the language, then redact PII in that language.
Run Desert Ant's Tongue to detect a message's language, then Redact to strip PII with the right language coverage, both on-device. Build it with the Desert Ant SDK. Tongue (Swift, Kotlin, JavaScript / TypeScript). Install and API: https://desertant.com/docs/tongue/. Redact (Swift, Kotlin, JavaScript / TypeScript). Install and API: https://desertant.com/docs/redact/. SDK source: https://github.com/Desert-Ant-Labs/desert-ant-core. Machine-readable catalog of every model and SDK: https://desertant.com/llms.txt.
Detect the language of a search query and pick the right index.
Run Desert Ant's Tongue on a search query on-device to route it to the right language index or analyzer, in microseconds. Build it with the Desert Ant SDK. Tongue (Swift, Kotlin, JavaScript / TypeScript). Install and API: https://desertant.com/docs/tongue/. Install: SwiftPM desert-ant-core; Maven ai.desertant:tongue:3.1.0; npm i @desert-ant-labs/tongue. SDK source: https://github.com/Desert-Ant-Labs/desert-ant-core. Machine-readable catalog of every model and SDK: https://desertant.com/llms.txt.
Build an OpenClaw skill that tags and language-routes text.
Write an OpenClaw skill that runs Desert Ant's Tongue to detect a text's language and Gist to tag its topics, on-device, so an agent can route content without a server call. Build it with the Desert Ant SDK. Tongue (Swift, Kotlin, JavaScript / TypeScript). Install and API: https://desertant.com/docs/tongue/. Gist (Swift, Kotlin, JavaScript / TypeScript). Install and API: https://desertant.com/docs/gist/. SDK source: https://github.com/Desert-Ant-Labs/desert-ant-core. Machine-readable catalog of every model and SDK: https://desertant.com/llms.txt.
A chat bot that tags and language-routes messages, in Node.
Build a Slack or Discord bot that runs Desert Ant's Tongue and Gist in Node to detect each message's language and topic and route it, with no third-party classification service. Build it with the Desert Ant SDK. Tongue (Swift, Kotlin, JavaScript / TypeScript). Install and API: https://desertant.com/docs/tongue/. Gist (Swift, Kotlin, JavaScript / TypeScript). Install and API: https://desertant.com/docs/gist/. SDK source: https://github.com/Desert-Ant-Labs/desert-ant-core. Machine-readable catalog of every model and SDK: https://desertant.com/llms.txt.
Build an agent skill that tags and language-routes text for an agent.
Write a SKILL.md agent skill that runs Desert Ant's Tongue to detect a text's language and Gist to tag its topics, on-device, so the agent can route or filter content without a server call. Build it with the Desert Ant SDK. Tongue (Swift, Kotlin, JavaScript / TypeScript). Install and API: https://desertant.com/docs/tongue/. Gist (Swift, Kotlin, JavaScript / TypeScript). Install and API: https://desertant.com/docs/gist/. SDK source: https://github.com/Desert-Ant-Labs/desert-ant-core. Machine-readable catalog of every model and SDK: https://desertant.com/llms.txt.
What the model does
- 84 languages: 59 decoded by the model, 25 settled by script alone.
- 2MB int8 weights, bundled in the package; nothing is downloaded.
- Tens of microseconds per detection, on the main thread.
- Identical answers on Swift, Kotlin, and JavaScript, held to shared golden vectors.
- Reports a reliability score and marks every tie.
Getting started
Add language identification to your iOS or macOS, Android or web app in a few lines of code. Tongue docs.
// Swift Package Manager
.package(url: "https://github.com/Desert-Ant-Labs/desert-ant-core", from: "3.1.0")
// target dependency
.product(name: "Tongue", package: "desert-ant-core")import Tongue
let tongue = try Tongue()
tongue.detect("kann ich das haben").language
// "de"Add Tongue from Desert Ant Labs to this Swift project (iOS, macOS). What it does: On-device Language Identification. SDK: Swift (iOS, macOS) Repo: https://github.com/Desert-Ant-Labs/desert-ant-core#readme // Swift Package Manager .package(url: "https://github.com/Desert-Ant-Labs/desert-ant-core", from: "3.1.0") // target dependency .product(name: "Tongue", package: "desert-ant-core") Reference: - Model page: https://desertant.com/models/tongue/ - Full catalog and other models: https://desertant.com/llms.txt Add the SDK, then follow its README for the exact API and current version. Do not invent API names or method signatures; confirm them against the README.
// build.gradle.kts (Maven Central)
implementation("ai.desertant:tongue:3.1.0")import ai.desertant.tongue.Tongue
val tongue = Tongue.bundled(context)
tongue.detect("kann ich das haben").language
// "de"Add Tongue from Desert Ant Labs to this Kotlin project (Android).
What it does: On-device Language Identification.
SDK:
Kotlin (Android)
Repo: https://github.com/Desert-Ant-Labs/desert-ant-core#readme
// build.gradle.kts (Maven Central)
implementation("ai.desertant:tongue:3.1.0")
Reference:
- Model page: https://desertant.com/models/tongue/
- Full catalog and other models: https://desertant.com/llms.txt
Add the SDK, then follow its README for the exact API and current version. Do not invent API names or method signatures; confirm them against the README.
npm i @desert-ant-labs/tongueimport { Tongue } from "@desert-ant-labs/tongue";
const tongue = await Tongue.load();
tongue.detect("kann ich das haben").language;
// "de"Add Tongue from Desert Ant Labs to this JavaScript / TypeScript project (Web, Node.js). What it does: On-device Language Identification. SDK: JavaScript / TypeScript (Web, Node.js) Repo: https://github.com/Desert-Ant-Labs/desert-ant-core#readme npm i @desert-ant-labs/tongue Reference: - Model page: https://desertant.com/models/tongue/ - Full catalog and other models: https://desertant.com/llms.txt Add the SDK, then follow its README for the exact API and current version. Do not invent API names or method signatures; confirm them against the README.
Specs
- Languages
- 84, across 31 scripts
- On-device size
- 2MB int8 weights, inside the package
- Speed
- Tens of microseconds per detection
- Model
- Script router plus hashed character n-grams into an int8 linear head
One or two words is often undecidable, and no model size fixes the problem. Tongue reports a tie or a tentative answer, so treat low reliability as "unknown", not as an answer.
Malay and Indonesian share too much vocabulary to separate on a short sample. Mongolian is covered in the traditional script only, so Cyrillic Mongolian is read as Russian.
Brand names, numbers, and code are not language, and will get a guess.
FAQ
What is Tongue?
On-device language detection for text across 84 languages. Identify the language in tens of microseconds, from just a few words.
Does Tongue run on device?
Yes. Tongue runs on the device, with no server call, so the data stays with the user.
Which platforms does Tongue support?
Tongue ships as a native on-device SDK for Swift, Kotlin, JavaScript / TypeScript.
How much does Tongue cost?
Every model is free up to 100k monthly active devices per SDK. Unlimited inference per user. Contact us for custom licenses.
How accurate or fast is Tongue?
At 2MB, Tongue scores 0.933 on three-word FLORES-200 input, against 0.887 for a 293MB detector.