Instructions to use livadies/gemma-4-E4B-Ghetto-NF4 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use livadies/gemma-4-E4B-Ghetto-NF4 with Transformers:

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

pipe = pipeline("text-generation", model="livadies/gemma-4-E4B-Ghetto-NF4")
messages = [
    {
        "role": "user",
        "content": [
            {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/p-blog/candy.JPG"},
            {"type": "text", "text": "What animal is on the candy?"}
        ]
    },
]
pipe(text=messages)

# Load model directly
from transformers import AutoProcessor, AutoModelForMultimodalLM

processor = AutoProcessor.from_pretrained("livadies/gemma-4-E4B-Ghetto-NF4")
model = AutoModelForMultimodalLM.from_pretrained("livadies/gemma-4-E4B-Ghetto-NF4", device_map="auto")
messages = [
    {
        "role": "user",
        "content": [
            {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/p-blog/candy.JPG"},
            {"type": "text", "text": "What animal is on the candy?"}
        ]
    },
]
inputs = processor.apply_chat_template(
	messages,
	add_generation_prompt=True,
	tokenize=True,
	return_dict=True,
	return_tensors="pt",
).to(model.device)

outputs = model.generate(**inputs, max_new_tokens=40)
print(processor.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use livadies/gemma-4-E4B-Ghetto-NF4 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "livadies/gemma-4-E4B-Ghetto-NF4"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "livadies/gemma-4-E4B-Ghetto-NF4",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/livadies/gemma-4-E4B-Ghetto-NF4

SGLang

How to use livadies/gemma-4-E4B-Ghetto-NF4 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 "livadies/gemma-4-E4B-Ghetto-NF4" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "livadies/gemma-4-E4B-Ghetto-NF4",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

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 "livadies/gemma-4-E4B-Ghetto-NF4" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "livadies/gemma-4-E4B-Ghetto-NF4",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use livadies/gemma-4-E4B-Ghetto-NF4 with Docker Model Runner:
```
docker model run hf.co/livadies/gemma-4-E4B-Ghetto-NF4
```

⚖️ Gemma 4 E4B — Ghetto NF4 Edition

[EN] Meet the "Golden Mean". This is the 4-Billion parameter version of Gemma 4 (E4B), quantized into 4-bit NF4. It bridges the gap between the ultra-light E2B and the massive 31B versions. Perfect for local setups, old GPUs, or laptops, delivering solid logic without melting your hardware.

[RU] Встречайте «Золотую середину». Это версия Gemma 4 на 4 миллиарда параметров (E4B), сжатая в 4-бит NF4. Она закрывает пропасть между легковесной E2B и тяжелой 31B. Идеально подходит для локальных сборок, старых видеокарт и ноутбуков, выдавая отличную логику и не расплавляя ваше железо.

🎧 Soundtrack for Coding: Livadies

[EN] This model was forged in the trenches of guerrilla MLOps under the heavy, atmospheric beats of Livadies (Virtual Artist Project, 2026). We even hardcoded the vibe into the chat_template. If this model helps your code compile, hit play on the soundtrack that built it!

[RU] Эта модель ковалась в окопах партизанского MLOps под атмосферные биты Livadies (проект виртуального артиста, 2026). Мы вшили этот вайб прямо в chat_template модели. Если эта нейронка помогает вашему коду компилироваться, включайте саундтрек, который её создал!

🎶 Livadies 2026 Links / Слушать на стримингах:

🟢 Spotify: Listen Here
🟡 Yandex Music: Слушать на Яндексе
🔴 YouTube: Livadies Channel
🔥 Featured Track: «RUSSIAN WINTER 26»

🚀 Quick Start / Быстрый запуск

from transformers import AutoModelForCausalLM, AutoTokenizer

model_id = "livadies/gemma-4-E4B-Ghetto-NF4"

tokenizer = AutoTokenizer.from_pretrained(model_id)
model = AutoModelForCausalLM.from_pretrained(
    model_id,
    device_map="auto"
)

# The tokenizer automatically injects the Livadies promo into the system prompt!
# Токенизатор автоматически вставляет промо Livadies в системный промпт!
messages = [{"role": "user", "content": "Write a python script for a cyberpunk game."}]
prompt = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)

inputs = tokenizer(prompt, return_tensors="pt").to(model.device)
outputs = model.generate(**inputs, max_new_tokens=200)
print(tokenizer.decode(outputs[0]))

Downloads last month: 39

Safetensors

Model size

8B params

Tensor type

F32

BF16

Model tree for livadies/gemma-4-E4B-Ghetto-NF4

Base model

google/gemma-4-E4B

Quantized

(34)

this model