A conversation, named as you type.

What it (weighs)

The benchmark, every language, every precision and the charts are on the model card →

In aparté, (three lines)

The titler plugin binds the model to the conversation manager: the first message of every conversation is titled by it, and the title lands in the list. The model loads once, the first time a title is needed.

import { loadTitler } from '@aparte/titler-latin';
import { setupTitler } from '@aparte/plugin-titler';

setupTitler(manager, { titler: loadTitler });   // manager: your AparteConversationManager

Without aparté, the runtime alone. One import, one file, one call:

import { loadTitler } from '@aparte/titler-latin';

const titler = await loadTitler();
titler.title("Peux-tu m'expliquer la photosynthèse chez les plantes ?");

npm install @aparte/titler-latin — the file ships inside the package and your bundler copies it as an asset, so the page fetches it from your own origin.

Where it (runs)

One model, one runtime per language. Each reads the same files and gives the same title, character for character.

JavaScript

Browsers, Node npm install @aparte/titler-latin

Swift

iOS, macOS, tvOS, watchOS, visionOS, Linux, Windows Swift Package Manager: https://github.com/apartejs/aparte-titler-swift

Kotlin

Android 5.0+, any JVM 8+ — from Kotlin or Java Maven Central: dev.apartejs:titler-latin

In Swift, AparteTitlerLatin brings the runtime and the 17-language model:

import AparteTitlerLatin

let titler = try AparteTitlerLatin.load()
titler.title("Peux-tu m'expliquer la photosynthèse chez les plantes ?")

In Xcode, File → Add Package Dependencies… and paste the repository URL. AparteTitler, the runtime alone, reads any model file from the model card.

In Kotlin, on Android or the JVM, TitlerLatin does the same:

import dev.apartejs.titler.latin.TitlerLatin

val titler = TitlerLatin.load()
titler.title("Peux-tu m'expliquer la photosynthèse chez les plantes ?")

Add dev.apartejs:titler-latin to your Gradle dependencies; it is usable from Java too. On Android, load it off the main thread. dev.apartejs:titler, the runtime alone, reads any model file from the model card.

Pick the (file)

One model per language group, four precisions each. The int3 files below ship on npm; every other one is on the model card.

What it (is), and is not

Extractive

It copies 3 to 6 words of the message, in the message’s order. It does not rephrase, summarise or translate — which is why it fits in 133 KB, and why a title reads like the message that started the conversation.

One message

Give it the user’s first message and nothing else — no system prompt, no history. It reads up to 1,200 characters; a long preamble before the actual request costs it some accuracy.

One file

Header, tokenizer and weights in a single .bin, read by a dependency-free runtime of about 250 lines. A Python reference reader gives the same title, word for word, on every message of the benchmark.

Open

MIT, code and weights. Training data and the benchmark are published beside the model, under their own licences.

The model card → The code → On npm → For Swift → For Kotlin → The plugin →