Instructions to use emese-tech/csermely-gguf with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use emese-tech/csermely-gguf with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="emese-tech/csermely-gguf",
	filename="csermely-q4_k_m.gguf",
)

output = llm(
	"Once upon a time,",
	max_tokens=512,
	echo=True
)
print(output)

Notebooks
Google Colab
Kaggle
Local Apps

llama.cpp

How to use emese-tech/csermely-gguf with llama.cpp:

Install from brew

brew install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf emese-tech/csermely-gguf:Q4_K_M
# Run inference directly in the terminal:
llama-cli -hf emese-tech/csermely-gguf:Q4_K_M

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf emese-tech/csermely-gguf:Q4_K_M
# Run inference directly in the terminal:
llama-cli -hf emese-tech/csermely-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 emese-tech/csermely-gguf:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf emese-tech/csermely-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 emese-tech/csermely-gguf:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf emese-tech/csermely-gguf:Q4_K_M

Use Docker

docker model run hf.co/emese-tech/csermely-gguf:Q4_K_M

LM Studio
Jan

vLLM

How to use emese-tech/csermely-gguf with vLLM:

Install from pip and serve model

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

Use Docker

docker model run hf.co/emese-tech/csermely-gguf:Q4_K_M

Ollama
How to use emese-tech/csermely-gguf with Ollama:
```
ollama run hf.co/emese-tech/csermely-gguf:Q4_K_M
```

Unsloth Studio new

How to use emese-tech/csermely-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 emese-tech/csermely-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 emese-tech/csermely-gguf to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for emese-tech/csermely-gguf to start chatting

Docker Model Runner
How to use emese-tech/csermely-gguf with Docker Model Runner:
```
docker model run hf.co/emese-tech/csermely-gguf:Q4_K_M
```

Lemonade

How to use emese-tech/csermely-gguf with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull emese-tech/csermely-gguf:Q4_K_M

Run and chat with the model

lemonade run user.csermely-gguf-Q4_K_M

List all available models

lemonade list

Csermely (GGUF)

GGUF quantized versions of Csermely — a 138M parameter Hungarian language model. Part of the Emese model family.

Compatible with llama.cpp, Ollama, LM Studio, and other GGUF-compatible runtimes.

For the full-precision HuggingFace version, see emese-tech/csermely.

Available Quantizations

File	Quantization	Size	Description
`csermely-q8_0.gguf`	Q8_0	173 MB	8-bit, near-lossless quality
`csermely-q4_k_m.gguf`	Q4_K_M	103 MB	4-bit, good quality/size balance

Usage

llama.cpp

./llama-cli -m csermely-q8_0.gguf -p "A magyar nyelv" -n 100 --repeat-penalty 1.2 --chat-template none

Ollama

ollama run emese-tech/csermely-gguf

Model Details


Parameters	137.8M
Architecture	LLaMA-style (decoder-only transformer)
Context length	8,192 tokens (YaRN RoPE)
Vocabulary	32,000 (SentencePiece Unigram, Hungarian)
License	MIT

Downloads last month: 178

GGUF

Model size

0.2B params

Architecture

llama

Hardware compatibility

4-bit

8-bit