dict-xs-mlc-q4f16_1

Ikhou Dictionary Model (dict-xs) - MLC Compiled (q4f16_1)

A lightweight multilingual dictionary model based on Qwen3-0.6B, fine-tuned on 1.7M dictionary-style glosses across 50+ languages.

This is the MLC-compiled version optimized for WebLLM/WebGPU inference with q4f16_1 quantization (~335 MB).

For the original PyTorch model, see: ikhou/dict-xs

Model Description

This model provides short, dictionary-style translations and glosses for words and phrases in context. It's designed for:

• Quick word lookups in reading applications
• Vocabulary learning tools
• Translation assistance
• Language learning applications

Key Features:

• 🌍 50+ languages supported (see list below)
• 📖 Dictionary-style glosses with grammatical markers
• ⚡ Ultra-fast inference (~335 MB with q4f16_1 quantization)
• 🎯 Context-aware translations
• 🌐 WebGPU/WebLLM ready - runs in browser
• 💾 Small footprint - ~75% smaller than original model

Supported Languages (50)

Major European Languages

English, German, French, Spanish, Italian, Portuguese, Dutch, Polish, Swedish, Danish, Norwegian Bokmål, Finnish, Czech, Romanian, Hungarian, Catalan, Greek

Cyrillic Script

Russian, Ukrainian, Bulgarian, Serbian

Asian Languages

Chinese (Mandarin), Japanese, Korean, Hindi, Bengali, Urdu, Tamil, Telugu, Marathi, Thai, Vietnamese, Indonesian, Malay, Filipino

Middle Eastern

Arabic, Persian, Turkish, Hebrew

African

Swahili, Amharic, Yoruba

Other

Lithuanian, Slovenian, Estonian, Latvian, Slovak, Croatian, Azerbaijani, Kazakh, Uzbek

Grammar Markers Explained

The model outputs grammatical information using standard linguistic abbreviations:

Noun Markers (Gender-based Languages)

• nm. = Masculine noun (e.g., "nm. roi, monarque" = king, monarch in French)
• nf. = Feminine noun (e.g., "nf. maison, demeure" = house, dwelling in French)
• nn. = Neuter noun (German, Russian) (e.g., "nn. Haus, Gebäude" = house, building)

Noun Markers (Non-gendered Languages)

• n. = Noun (e.g., "n. house, home" in English)

Other Parts of Speech

• adj. = Adjective (e.g., "adj. rapide, vite" = fast, quick)
• adv. = Adverb (e.g., "adv. rapidement, vite" = quickly, fast)
• pp = Past participle (e.g., "mangé → eaten, consumed (pp)")

Verb Forms

For conjugated verbs, the model provides:

• Translation(s)
• Tense/mood information in parentheses
• Example: "venait → came, was coming (imparfait, il)" = imperfect tense, he

Usage

⚠️ Important Note

WebGPU Library Status: The weights have been updated, but the WebGPU wasm file needs to be recompiled. Until then, please use one of these alternatives:

1. MLC-LLM CLI (recommended - works immediately)
2. Original PyTorch model at ikhou/dict-xs
3. Compile your own wasm (see instructions at bottom of README)

MLC-LLM CLI Usage (Recommended)

# Install mlc-llm
pip install --pre -U -f https://mlc.ai/wheels mlc-llm-nightly-cpu mlc-ai-nightly-cpu

# Run the model
mlc_llm chat HF://ikhou/dict-xs-mlc-q4f16_1

WebLLM Usage (Browser - Requires Wasm Recompilation)

This model is optimized for WebLLM and can run directly in the browser once the WebGPU library is recompiled:

import * as webllm from "@mlc-ai/web-llm";

const messages = [
  {
    role: "system",
    content: "You are a bilingual dictionary. Given a word/phrase in context, output a short gloss.\n\nRules:\n- One line only, no labels\n- Use grammar markers: nm./nf./nn. for gendered nouns, n. for others, adj., adv., verbs with tense info\n- 1-4 short translations, comma-separated\n- Apply markers based on definition language"
  },
  {
    role: "user",
    content: 'Expression: "maison"\nContext: Il habite dans une petite 【maison】 près de la mer.\nSource language: fra (French)\nDefinition language: eng (English)\n\nReturn the single-line gloss now.'
  }
];

const engine = await webllm.CreateMLCEngine(
  "ikhou/dict-xs-mlc-q4f16_1",
  {
    initProgressCallback: (progress) => {
      console.log(progress);
    }
  }
);

const reply = await engine.chat.completions.create({
  messages,
  max_tokens: 50,
  temperature: 0.3,
});

console.log(reply.choices[0].message.content); // Output: "nf. house, home"

Note: The above code will load but return empty outputs until the wasm is recompiled.

Original PyTorch Usage

For the original model, see ikhou/dict-xs:

from transformers import AutoModelForCausalLM, AutoTokenizer

model = AutoModelForCausalLM.from_pretrained(
    "ikhou/dict-xs",
    torch_dtype="bfloat16",
    device_map="auto"
)
tokenizer = AutoTokenizer.from_pretrained("ikhou/dict-xs")

# Example: Get a French→English gloss
messages = [
    {
        "role": "system",
        "content": "You are a bilingual dictionary. Given a word/phrase in context, output a short gloss.\n\nRules:\n- One line only, no labels\n- Use grammar markers: nm./nf./nn. for gendered nouns, n. for others, adj., adv., verbs with tense info\n- 1-4 short translations, comma-separated\n- Apply markers based on definition language"
    },
    {
        "role": "user",
        "content": 'Expression: "maison"\nContext: Il habite dans une petite 【maison】 près de la mer.\nSource language: fra (French)\nDefinition language: eng (English)\n\nReturn the single-line gloss now.'
    }
]

inputs = tokenizer.apply_chat_template(
    messages,
    add_generation_prompt=True,
    return_tensors="pt"
).to(model.device)

outputs = model.generate(
    inputs,
    max_new_tokens=50,
    temperature=0.3,
    do_sample=True,
    top_p=0.9
)

response = tokenizer.decode(outputs[0][inputs.shape[1]:], skip_special_tokens=True)
print(response)  # Output: "nf. house, home"

Input Format

The model expects:

1. Expression: The word/phrase to define
2. Context: Sentence with the expression (use 【】 to highlight)
3. Source language: ISO 639-3 code (e.g., fra, eng, deu)
4. Definition language: ISO 639-3 code

Output Format

The model returns a single line with:

• Grammar marker (nm./nf./nn./n./adj./adv./pp)
• 1-4 short translations/synonyms, comma-separated
• For verbs: glosses + grammatical info in parentheses

Training Details

Training Data

• Dataset: 1.7M synthetic dictionary entries
• Sources: FineWeb (English), FineWeb-2 (49 other languages)
• Generation: GPT-4-based teacher model for quality glosses
• Filtering: Proper noun filtering, quality scoring

Training Configuration

• Base Model: Qwen/Qwen3-0.6B
• Training Type: Full fine-tuning (not LoRA)
• Precision: bfloat16
• Batch Size: 32 per device
• Gradient Accumulation: 8 steps
• Total Steps: 6,568
• Learning Rate: AdamW with cosine schedule
• Hardware: NVIDIA H100 (95GB)
• Training Time: ~6 hours

Training Results

• Final Loss: 1.30
• Eval Loss: 1.34
• Training thoroughly validated - no zero loss issues

Model Architecture

• Architecture: Qwen3ForCausalLM
• Parameters: 596M (original), ~335 MB (quantized)
• Layers: 28 transformer layers
• Hidden Size: 1024
• Attention Heads: 16 (8 KV heads)
• Context Length: 40,960 tokens (model max, trained on 512)
• Vocabulary: 151,936 tokens (151,669 active)

MLC Compilation Details

• Quantization: q4f16_1 (4-bit weights, float16 activations)
• Group Size: 32
• Compiled With: MLC-LLM 0.20.dev107
• Target: WebGPU/CPU
• Size Reduction: ~75% (1.2 GB → 335 MB)

Limitations

• Context: Works best with clear, simple contexts
• Proper nouns: May struggle with names, places, brands
• Rare languages: Better performance on high-resource languages
• Multi-word phrases: Best for 1-6 token phrases
• Ambiguity: Provides common meanings, may miss context-specific nuances

Ethical Considerations

• Bias: Trained on web data which may contain biases
• Not for sensitive applications: Dictionary glosses may have errors
• Educational use: Best for learning and reference, not authoritative translation

Compiling WebGPU Library (Advanced)

To compile the WebGPU wasm file for browser use:

Prerequisites

1. Install Emscripten:

git clone https://github.com/emscripten-core/emsdk.git
cd emsdk
./emsdk install latest
./emsdk activate latest
source ./emsdk_env.sh

1. Build MLC-LLM from source:

git clone --recursive https://github.com/mlc-ai/mlc-llm.git
cd mlc-llm
./web/prep_emcc_deps.sh
export MLC_LLM_SOURCE_DIR=$(pwd)

Compile for WebGPU

# Download this model's config
git clone https://huggingface.co/ikhou/dict-xs-mlc-q4f16_1

# Compile
python -m mlc_llm compile dict-xs-mlc-q4f16_1/mlc-chat-config.json \
    --device webgpu \
    --opt O3 \
    -o dict-xs-q4f16_1-webgpu.wasm

# The compiled wasm can then be uploaded to replace the old one

License

Apache 2.0

Citation

@misc{ikhou-dict-xs,
  author = {Ikhou},
  title = {Ikhou Dictionary Model (dict-xs)},
  year = {2026},
  publisher = {Hugging Face},
  howpublished = {\url{https://huggingface.co/ikhou/dict-xs}}
}

Acknowledgments

• Based on Qwen3-0.6B by Alibaba Cloud
• Training data sourced from FineWeb and FineWeb-2
• Trained with Hugging Face Transformers

Contact

For issues or questions, please open an issue on the model repository.