Instructions to use domofon/domofon-0.8b-75b-tokens-base-GGUF with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use domofon/domofon-0.8b-75b-tokens-base-GGUF with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="domofon/domofon-0.8b-75b-tokens-base-GGUF",
	filename="domofon-0.8b-75b-tokens-base-Q4_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 domofon/domofon-0.8b-75b-tokens-base-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 domofon/domofon-0.8b-75b-tokens-base-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf domofon/domofon-0.8b-75b-tokens-base-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 domofon/domofon-0.8b-75b-tokens-base-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf domofon/domofon-0.8b-75b-tokens-base-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 domofon/domofon-0.8b-75b-tokens-base-GGUF:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf domofon/domofon-0.8b-75b-tokens-base-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 domofon/domofon-0.8b-75b-tokens-base-GGUF:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf domofon/domofon-0.8b-75b-tokens-base-GGUF:Q4_K_M

Use Docker

docker model run hf.co/domofon/domofon-0.8b-75b-tokens-base-GGUF:Q4_K_M

LM Studio
Jan

vLLM

How to use domofon/domofon-0.8b-75b-tokens-base-GGUF with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "domofon/domofon-0.8b-75b-tokens-base-GGUF"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "domofon/domofon-0.8b-75b-tokens-base-GGUF",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/domofon/domofon-0.8b-75b-tokens-base-GGUF:Q4_K_M

Ollama
How to use domofon/domofon-0.8b-75b-tokens-base-GGUF with Ollama:
```
ollama run hf.co/domofon/domofon-0.8b-75b-tokens-base-GGUF:Q4_K_M
```

Unsloth Studio

How to use domofon/domofon-0.8b-75b-tokens-base-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 domofon/domofon-0.8b-75b-tokens-base-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 domofon/domofon-0.8b-75b-tokens-base-GGUF to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for domofon/domofon-0.8b-75b-tokens-base-GGUF to start chatting

Atomic Chat new
Docker Model Runner
How to use domofon/domofon-0.8b-75b-tokens-base-GGUF with Docker Model Runner:
```
docker model run hf.co/domofon/domofon-0.8b-75b-tokens-base-GGUF:Q4_K_M
```

Lemonade

How to use domofon/domofon-0.8b-75b-tokens-base-GGUF with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull domofon/domofon-0.8b-75b-tokens-base-GGUF:Q4_K_M

Run and chat with the model

lemonade run user.domofon-0.8b-75b-tokens-base-GGUF-Q4_K_M

List all available models

lemonade list

domofon-0.8b-75b-tokens-base-GGUF

это тестовый прогон модели на отметке 75B tokens pretraining. модель лучше всего понимает русский язык,

не содержит SFT chat прогона. Только base pretraining.

спонсировано Google TRC Research Cloud.

This research was supported with Cloud TPUs from Google's TPU Research Cloud (TRC).

Model details

Type: base causal language model
Parameters: 0.8B class, approximately 0.88B
Checkpoint: 77800
Training stage: base pretraining, before chat/SFT
Tokenizer vocab size: 248,072
Context used during training: 4096
Architecture family: Qwen3-style dense GQA
Weights: trained from scratch; upstream Qwen references are for architecture/tokenizer compatibility only

Usage with llama.cpp

llama-cli \
  -m domofon-0.8b-75b-tokens-base-Q4_K_M.gguf \
  -p "Домофон — это" \
  -n 128 \
  --temp 0.7 \
  --top-p 0.9

Usage with Ollama

ollama create domofon-base -f Modelfile.ollama
ollama run domofon-base

Notes

This is not the chat-SFT experiment. Do not expect instruction following, assistant-role formatting, or reliable multi-turn chat behavior from this base checkpoint. Use it for base-model probing, continuation tests, or as a starting point for later SFT/alignment experiments.

Downloads last month: 19

GGUF

Model size

0.9B params

Architecture

qwen3

Hardware compatibility

4-bit

Collection including domofon/domofon-0.8b-75b-tokens-base-GGUF

Domofon-temp

Collection

Temporary models of Domofon's model. (Un usable to real launch) • 3 items • Updated May 23