Instructions to use Abduqodir06/Lyra-Uz with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Abduqodir06/Lyra-Uz with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("text-generation", model="Abduqodir06/Lyra-Uz")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("Abduqodir06/Lyra-Uz")
model = AutoModelForCausalLM.from_pretrained("Abduqodir06/Lyra-Uz")
messages = [
    {"role": "user", "content": "Who are you?"},
]
inputs = tokenizer.apply_chat_template(
	messages,
	add_generation_prompt=True,
	tokenize=True,
	return_dict=True,
	return_tensors="pt",
).to(model.device)

outputs = model.generate(**inputs, max_new_tokens=40)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Inference
Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use Abduqodir06/Lyra-Uz with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Abduqodir06/Lyra-Uz"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Abduqodir06/Lyra-Uz",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/Abduqodir06/Lyra-Uz

SGLang

How to use Abduqodir06/Lyra-Uz with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "Abduqodir06/Lyra-Uz" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Abduqodir06/Lyra-Uz",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "Abduqodir06/Lyra-Uz" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Abduqodir06/Lyra-Uz",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use Abduqodir06/Lyra-Uz with Docker Model Runner:
```
docker model run hf.co/Abduqodir06/Lyra-Uz
```
Browse Quantizations to use this model in llama.cpp, Ollama, LM Studio, or any compatible app.

LYRA-Uz

LYRA-Uz — o‘zbek tilida yuqori sifatli ko‘rsatmalarni bajaruvchi (instruction-tuned) ochiq manbali til modeli. Mistral-7B-Instruct-v0.3 arxitekturasi asosida o‘zbek va ingliz tillaridagi maʼlumotlar bilan maxsus oʻqitilgan. Apache 2.0 litsenziyasi bilan erkin foydalanish mumkin.

Asosiy imkoniyatlari

Savol-javob — o‘zbek tilidagi umumiy bilim savollariga javob beradi
Matnni umumlashtirish — uzoq matnlarni qisqacha bayon qiladi
Tarjima — o‘zbek ↔ ingliz tillari o‘rtasida tarjima qiladi
Matn tasnifi — yangiliklar, hissiy tahlil va boshqa toifalarga ajratadi
Ko‘rsatmalarni bajarish — berilgan vazifani ingliz va o‘zbek tillarida tushunib, bajaradi

Tezkor ishlatish



```python
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("Abduqodir06/Lyra-Uz")
model = AutoModelForCausalLM.from_pretrained(
    "Abduqodir06/Lyra-Uz",
    load_in_4bit=True,
    device_map="auto"
)

prompt = "O'zbekiston poytaxti qaysi?"
inputs = tokenizer(prompt, return_tensors="pt").to("cuda")
outputs = model.generate(**inputs, max_new_tokens=128)
print(tokenizer.decode(outputs[0], skip_special_tokens=True))

LYRA loyihasi haqida

Ushbu model LYRA (Large Uzbek Language Reasoning Architecture) loyihasining bir qismidir. To‘liq loyiha quyidagi bosqichlarni o‘z ichiga oladi (hozirda ustida ish olib borilmoqda):

Tokenizer optimallashtirish — o‘zbek tili morfologiyasiga mos BPE
RAG va veb-qidiruv — bilimlarni real vaqtda qidirib javob berish
Deploy — FastAPI, Telegram bot va ommaviy foydalanish

Hozirgi holat: Ushbu repo’dagi model LYRA loyihasining birinchi tayyor komponentidir. Qo‘shimcha funksiyalar (RAG, veb-qidiruv, maxsus tokenizator) ustida ish olib borilmoqda.

Texnik maʼlumotlar

Xususiyat	Qiymat
Parametrlar soni	7 milliard
Arxitektura	Mistral-7B-Instruct-v0.3
Litsenziya	Apache 2.0
GPU talabi (FP16)	~14.5 GB VRAM
GPU talabi (4-bit)	~4.5 GB VRAM
Tillari	o‘zbek (asosiy), ingliz

Downloads last month: 1,990

Model tree for Abduqodir06/Lyra-Uz

Base model

mistralai/Mistral-7B-v0.3

Finetuned

mistralai/Mistral-7B-Instruct-v0.3

Finetuned

(494)

this model

Quantizations

1 model