Qwen3-14B-Base-Uzbek-Cyrillic 🇺🇿

This model is a fine-tuned variant of Qwen3-14B-Base, adapted for the Uzbek language (Cyrillic script) using LoRA and Unsloth.

It retains Qwen3’s powerful reasoning, comprehension, and multilingual understanding — while specializing in Uzbek Cyrillic vocabulary, syntax, and cultural nuance.

---

🚀 Model Overview

Property	Value
Base model	Qwen/Qwen3-14B-Base
Architecture	Transformer Decoder (Causal LM)
Parameters	14.8B
Context length	32,768 tokens
Finetuning method	LoRA (r=16, α=32, dropout=0.0)
Training framework	Unsloth
Precision	bfloat16
Languages	Uzbek (Cyrillic), multilingual

---

🧠 Purpose

This model is designed to:

• Generate natural and grammatically correct Uzbek Cyrillic text
• Support content generation, chat, and summarization in Uzbek
• Serve as a strong multilingual foundation for Central Asian applications

---

🧩 Usage

You can load and run it directly using transformers or vLLM.

Transformers Example

from transformers import AutoModelForCausalLM, AutoTokenizer
import torch

model_id = "Just-Bax/Qwen3-14B-Base-Uzbek-Cyrillic"

tokenizer = AutoTokenizer.from_pretrained(model_id)
model = AutoModelForCausalLM.from_pretrained(model_id, torch_dtype=torch.bfloat16, device_map="auto")

prompt = "Ассалому алайкум! Бугунги кун ҳақида маълумот беринг."
inputs = tokenizer(prompt, return_tensors="pt").to(model.device)
outputs = model.generate(**inputs, max_new_tokens=100)
print(tokenizer.decode(outputs[0], skip_special_tokens=True))