Instructions to use lerugray/ferney-7b with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use lerugray/ferney-7b with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="lerugray/ferney-7b",
	filename="ferney-7b-Q5_K_M.gguf",
)

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

Notebooks
Google Colab
Kaggle
Local Apps Settings

llama.cpp

How to use lerugray/ferney-7b 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 lerugray/ferney-7b:Q5_K_M
# Run inference directly in the terminal:
llama cli -hf lerugray/ferney-7b:Q5_K_M

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf lerugray/ferney-7b:Q5_K_M
# Run inference directly in the terminal:
llama cli -hf lerugray/ferney-7b:Q5_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 lerugray/ferney-7b:Q5_K_M
# Run inference directly in the terminal:
./llama-cli -hf lerugray/ferney-7b:Q5_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 lerugray/ferney-7b:Q5_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf lerugray/ferney-7b:Q5_K_M

Use Docker

docker model run hf.co/lerugray/ferney-7b:Q5_K_M

LM Studio
Jan

vLLM

How to use lerugray/ferney-7b with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "lerugray/ferney-7b"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "lerugray/ferney-7b",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/lerugray/ferney-7b:Q5_K_M

Ollama
How to use lerugray/ferney-7b with Ollama:
```
ollama run hf.co/lerugray/ferney-7b:Q5_K_M
```

Unsloth Studio

How to use lerugray/ferney-7b 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 lerugray/ferney-7b 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 lerugray/ferney-7b to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for lerugray/ferney-7b to start chatting

Atomic Chat new
Docker Model Runner
How to use lerugray/ferney-7b with Docker Model Runner:
```
docker model run hf.co/lerugray/ferney-7b:Q5_K_M
```

Lemonade

How to use lerugray/ferney-7b with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull lerugray/ferney-7b:Q5_K_M

Run and chat with the model

lemonade run user.ferney-7b-Q5_K_M

List all available models

lemonade list

ferney: Voltaire (François-Marie Arouet, 1694–1778)

VOICE MODEL, the author's own recorded prose. This model trains on Voltaire's own words in public-domain English translation. It is a small model tuned to hold a voice, and it invents freely, so read it for the register, not the record.

Part of the Elect voice-model series: historical figures as runnable instruments. Voltaire is named here for Ferney, the village on the French-Swiss border where he settled and wrote in his later years.

What this is

A fine-tuned Qwen2.5-7B-Instruct that targets the register of Voltaire: the wit, the mockery of dogma and superstition, the plea for tolerance, the dry contempt for cruelty dressed up as piety. It speaks in the first person, in the manner of the philosophe who spent his life setting reason against fanaticism.

The model is a full fine-tune distilled to Voltaire's first-person register and register-purity-filtered so it speaks as Voltaire rather than narrating about him.

The source

The corpus is Voltaire's own prose, in public-domain English translation:

the Philosophical Dictionary (the William F. Fleming translation)
the Letters on England (also titled the Philosophical Letters / Letters Concerning the English Nation)
the Treatise on Toleration

The build keeps Voltaire's voice and drops the translators' and editors' apparatus (prefaces, footnotes, third-person framing), filtering the text down to his first-person register. The corpus is not bundled with these weights; it is rebuildable from the cited public-domain sources.

Example exchanges

Real outputs from this model:

What is philosophy? We take opium; we call it philosophy.

What do you think of priests? Sir, I am persuaded that without your priests there will be fewer murders.

What is war? War is a crime committed by order of the legislature.

Running it

Q5_K_M GGUF, served with Ollama. Pair it with the elicitation frame (a "visitor asks Voltaire" template); see the Modelfile in the project repo. Plain chat templates mis-wrap the completion register and produce rambling, so the completion frame keeps it in voice.

Intended uses

The Elect console (Ollama serve plus chat)
Enlightenment-register dialogue and character embodiment
Creative and educational use; a wit-and-tolerance sparring partner

Limitations

This is a 7B model, not a search index over Voltaire. It invents freely: names, dates, quotations, sources, events. It imitates the manner of Voltaire, the epigram and the ironic thrust, but the specific claims it makes are not reliable history and the quotations it attributes are not to be trusted. Treat every output as fiction in his voice, never as the record, and check anything before you repeat it.

License

CC-BY-NC-4.0. The model weights (GGUF) are released publicly under this license and are downloadable from this repository. The training corpus is not bundled: it is rebuildable from the cited public-domain sources (the Fleming translation of the Philosophical Dictionary, the Letters on England, and the Treatise on Toleration).

Part of the Elect, a roster of public-domain voice and register models. Ferney is the Enlightenment member.