Instructions to use Vaultek/Quartz-R1-8B-Genesis-GGUF with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Vaultek/Quartz-R1-8B-Genesis-GGUF with Transformers:

# Load model directly
from transformers import AutoModel
model = AutoModel.from_pretrained("Vaultek/Quartz-R1-8B-Genesis-GGUF", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

llama.cpp

How to use Vaultek/Quartz-R1-8B-Genesis-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 Vaultek/Quartz-R1-8B-Genesis-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf Vaultek/Quartz-R1-8B-Genesis-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 Vaultek/Quartz-R1-8B-Genesis-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf Vaultek/Quartz-R1-8B-Genesis-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 Vaultek/Quartz-R1-8B-Genesis-GGUF:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf Vaultek/Quartz-R1-8B-Genesis-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 Vaultek/Quartz-R1-8B-Genesis-GGUF:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf Vaultek/Quartz-R1-8B-Genesis-GGUF:Q4_K_M

Use Docker

docker model run hf.co/Vaultek/Quartz-R1-8B-Genesis-GGUF:Q4_K_M

LM Studio
Jan
Ollama
How to use Vaultek/Quartz-R1-8B-Genesis-GGUF with Ollama:
```
ollama run hf.co/Vaultek/Quartz-R1-8B-Genesis-GGUF:Q4_K_M
```

Unsloth Studio

How to use Vaultek/Quartz-R1-8B-Genesis-GGUF with Unsloth Studio:

Install Unsloth Studio (macOS, Linux, WSL)

curl -fsSL https://unsloth.ai/install.sh | sh
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for Vaultek/Quartz-R1-8B-Genesis-GGUF to start chatting

Install Unsloth Studio (Windows)

irm https://unsloth.ai/install.ps1 | iex
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for Vaultek/Quartz-R1-8B-Genesis-GGUF to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for Vaultek/Quartz-R1-8B-Genesis-GGUF to start chatting

Docker Model Runner
How to use Vaultek/Quartz-R1-8B-Genesis-GGUF with Docker Model Runner:
```
docker model run hf.co/Vaultek/Quartz-R1-8B-Genesis-GGUF:Q4_K_M
```

Lemonade

How to use Vaultek/Quartz-R1-8B-Genesis-GGUF with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull Vaultek/Quartz-R1-8B-Genesis-GGUF:Q4_K_M

Run and chat with the model

lemonade run user.Quartz-R1-8B-Genesis-GGUF-Q4_K_M

List all available models

lemonade list

Atomic Chat

Quartz-R1-8B-Genesis-GGUF / README.md

Vaultek

Create README.md

961d240 verified 1 day ago

preview code

Raw

History Blame Contribute Delete

2.43 kB

	---
	base_model:
	- Vaultek/Quartz-R1-8B-Genesis
	tags:
	- GGUF
	language:
	- ru
	- en
	library_name: transformers
	---

	Квантизованная версия Quartz-R1-8B-Genesis. Информация о модели доступна в основном репозитории: [`Quartz-R1-8B-Genesis`](https://huggingface.co/Vaultek/Quartz-R1-8B-Genesis/).

	# Использование с помощью llama.cpp
	Для начала нужно собрать [llama.cpp](https://github.com/ggml-org/llama.cpp) (или обновить, если уже есть):
	```bash
	git clone https://github.com/ggml-org/llama.cpp.git
	cd llama.cpp
	cmake -B build
	cmake --build build --config Release -j$(nproc)
	cd ..
	```

	Запуск модели в интерактивном режиме:
	```bash
	llama.cpp/build/bin/llama-cli -m quartz_r1_genesis_clean.Q4_K_M.gguf -c 32768 -np 4
	```

	Запуск сервера:
	```bash
	llama.cpp/build/bin/llama-server -m quartz_r1_genesis_clean.Q4_K_M.gguf -c 32768 -np 4
	```
	Если позволяют ресурсы, можно ускорить инференс, добавив `-t 10`.

	# Квантизации и их качество

	Вот созданный ikawrakow график, сравнивающий квантизации (на его странице, но данные совпадают)
	types (lower is better):

	![image.png](https://www.nethype.de/huggingface_embed/quantpplgraph.png)

	And here are Artefact2's thoughts on the matter:
	https://gist.github.com/Artefact2/b5f810600771265fc1e39442288e8ec9

	\| Link \| Type \| Size/GB \|
	\|:-----\|:-----\|--------:\|
	\| [GGUF_Q2_K](https://huggingface.co/Vaultek/Quartz-R1-8B-Genesis-GGUF/resolve/main/quartz_r1_genesis_clean.Q2_K.gguf) \| Q2_K \| 3.18 \|
	\| [GGUF_Q3_K_M](https://huggingface.co/Vaultek/Quartz-R1-8B-Genesis-GGUF/resolve/main/quartz_r1_genesis_clean.Q3_K_M.gguf) \| Q3_K_M \| 4.02 \|
	\| [GGUF_Q4_K_M](https://huggingface.co/Vaultek/Quartz-R1-8B-Genesis-GGUF/resolve/main/quartz_r1_genesis_clean.Q4_K_M.gguf) \| Q4_K_M \| 4.92 \|
	\| [GGUF_Q5_K_M](https://huggingface.co/Vaultek/Quartz-R1-8B-Genesis-GGUF/resolve/main/quartz_r1_genesis_clean.Q5_K_M.gguf) \| Q5_K_M \| 5.73 \|
	\| [GGUF_Q8_0](https://huggingface.co/Vaultek/Quartz-R1-8B-Genesis-GGUF/resolve/main/quartz_r1_genesis_clean.Q8_0.gguf) \| Q8_0 \| 8.54 \|
	\| [GGUF_F16](https://huggingface.co/mradermacher/Quartz-R1-8B-Genesis-GGUF/resolve/main/Quartz-R1-8B-Genesis.f16.gguf) \| f16 \| 16.2 \|