Instructions to use Lev384501/qwen3-0.6b-russian-dialogues-Q4_K_M-GGUF with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Lev384501/qwen3-0.6b-russian-dialogues-Q4_K_M-GGUF with Transformers:

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

pipe = pipeline("text-generation", model="Lev384501/qwen3-0.6b-russian-dialogues-Q4_K_M-GGUF")

# pip install -U transformers accelerate
# Load model directly
from transformers import AutoModel
model = AutoModel.from_pretrained("Lev384501/qwen3-0.6b-russian-dialogues-Q4_K_M-GGUF", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

llama.cpp

How to use Lev384501/qwen3-0.6b-russian-dialogues-Q4_K_M-GGUF with llama.cpp:

Install (macOS, Linux)

curl -LsSf https://llama.app/install.sh | sh
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf Lev384501/qwen3-0.6b-russian-dialogues-Q4_K_M-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf Lev384501/qwen3-0.6b-russian-dialogues-Q4_K_M-GGUF:Q4_K_M

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf Lev384501/qwen3-0.6b-russian-dialogues-Q4_K_M-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf Lev384501/qwen3-0.6b-russian-dialogues-Q4_K_M-GGUF:Q4_K_M

Use pre-built binary

# Download pre-built binary from:
# https://github.com/ggerganov/llama.cpp/releases
# Start a local OpenAI-compatible server with a web UI:
./llama-server -hf Lev384501/qwen3-0.6b-russian-dialogues-Q4_K_M-GGUF:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf Lev384501/qwen3-0.6b-russian-dialogues-Q4_K_M-GGUF:Q4_K_M

Build from source code

git clone https://github.com/ggerganov/llama.cpp.git
cd llama.cpp
cmake -B build
cmake --build build -j --target llama-server llama-cli
# Start a local OpenAI-compatible server with a web UI:
./build/bin/llama-server -hf Lev384501/qwen3-0.6b-russian-dialogues-Q4_K_M-GGUF:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf Lev384501/qwen3-0.6b-russian-dialogues-Q4_K_M-GGUF:Q4_K_M

Use Docker

docker model run hf.co/Lev384501/qwen3-0.6b-russian-dialogues-Q4_K_M-GGUF:Q4_K_M

Llama new
LM Studio
Jan

vLLM

How to use Lev384501/qwen3-0.6b-russian-dialogues-Q4_K_M-GGUF with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Lev384501/qwen3-0.6b-russian-dialogues-Q4_K_M-GGUF"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Lev384501/qwen3-0.6b-russian-dialogues-Q4_K_M-GGUF",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/Lev384501/qwen3-0.6b-russian-dialogues-Q4_K_M-GGUF:Q4_K_M

SGLang

How to use Lev384501/qwen3-0.6b-russian-dialogues-Q4_K_M-GGUF 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 "Lev384501/qwen3-0.6b-russian-dialogues-Q4_K_M-GGUF" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Lev384501/qwen3-0.6b-russian-dialogues-Q4_K_M-GGUF",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

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 "Lev384501/qwen3-0.6b-russian-dialogues-Q4_K_M-GGUF" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Lev384501/qwen3-0.6b-russian-dialogues-Q4_K_M-GGUF",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Ollama
How to use Lev384501/qwen3-0.6b-russian-dialogues-Q4_K_M-GGUF with Ollama:
```
ollama run hf.co/Lev384501/qwen3-0.6b-russian-dialogues-Q4_K_M-GGUF:Q4_K_M
```
Unsloth Desktop
Docker Model Runner
How to use Lev384501/qwen3-0.6b-russian-dialogues-Q4_K_M-GGUF with Docker Model Runner:
```
docker model run hf.co/Lev384501/qwen3-0.6b-russian-dialogues-Q4_K_M-GGUF:Q4_K_M
```

Lemonade

How to use Lev384501/qwen3-0.6b-russian-dialogues-Q4_K_M-GGUF with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull Lev384501/qwen3-0.6b-russian-dialogues-Q4_K_M-GGUF:Q4_K_M

Run and chat with the model

lemonade run user.qwen3-0.6b-russian-dialogues-Q4_K_M-GGUF-Q4_K_M

List all available models

lemonade list

Atomic Chat

qwen3-0.6b-russian-dialogues-Q4_K_M-GGUF

Квантование Q4_K_M модели ya-yje-krasni/qwen3-0.6b-russian-dialogues — полного файнтюнинга Qwen3-0.6B на русских диалогах (обучались все параметры, без LoRA и адаптеров).

Конвертировано в GGUF через llama.cpp с помощью GGUF-my-repo.


Квантование	Q4_K_M
Размер файла	397 МБ
Исходная модель	ya-yje-krasni/qwen3-0.6b-russian-dialogues (float32, 2.4 ГБ)
Базовая модель	Qwen/Qwen3-0.6B
Датасет	Den4ikAI/russian_dialogues
Контекст	32 768 токенов
Язык	русский
Лицензия	Apache 2.0

Более точная версия: Q8_0 (639 МБ).

Формат промпта

Модель обучена на шаблоне с явным маркером ответа, а не на chat template:

<реплика пользователя>
### Ответ:

Генерация продолжает текст после ### Ответ:. Поэтому используйте обычное дополнение текста (llama-cli, /completion), а не чат-эндпоинт.

Запуск через llama.cpp

Установка (macOS и Linux):

brew install llama.cpp

Одиночная генерация напрямую из Hub:

llama-cli --hf-repo Lev384501/qwen3-0.6b-russian-dialogues-Q4_K_M-GGUF \
  --hf-file qwen3-0.6b-russian-dialogues-q4_k_m.gguf \
  -p $'Привет, как дела?\n### Ответ:' -n 128 --no-display-prompt

Локальный сервер с OpenAI-совместимым API:

llama-server --hf-repo Lev384501/qwen3-0.6b-russian-dialogues-Q4_K_M-GGUF \
  --hf-file qwen3-0.6b-russian-dialogues-q4_k_m.gguf -c 4096

Скачать файл заранее:

pip install -U "huggingface_hub[cli]"
hf download Lev384501/qwen3-0.6b-russian-dialogues-Q4_K_M-GGUF \
  qwen3-0.6b-russian-dialogues-q4_k_m.gguf --local-dir .

Код и скрипты

Готовые скрипты запуска и конфиги: github.com/lev73748/qwen3-0.6b-russian-dialogues

Ограничения

Модель на 0.6B параметров плюс 4-битное квантование — ответы короткие, разговорные, факты могут быть неверными. Не подходит для задач, требующих точности.

Downloads last month: 169

GGUF

Model size

0.6B params

Architecture

qwen3

Hardware compatibility

4-bit

Model tree for Lev384501/qwen3-0.6b-russian-dialogues-Q4_K_M-GGUF

Base model

Qwen/Qwen3-0.6B-Base

Finetuned

Qwen/Qwen3-0.6B

Finetuned

ya-yje-krasni/qwen3-0.6b-russian-dialogues

Quantized

(2)

this model

Lev384501
/

qwen3-0.6b-russian-dialogues-Q4_K_M-GGUF