Instructions to use st4system/gemma4-e2b-scn-ollama-q4km with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use st4system/gemma4-e2b-scn-ollama-q4km with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="st4system/gemma4-e2b-scn-ollama-q4km",
	filename="gemma4-e2b-scn-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 st4system/gemma4-e2b-scn-ollama-q4km with llama.cpp:

Install from brew

brew install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf st4system/gemma4-e2b-scn-ollama-q4km:Q4_K_M
# Run inference directly in the terminal:
llama-cli -hf st4system/gemma4-e2b-scn-ollama-q4km:Q4_K_M

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf st4system/gemma4-e2b-scn-ollama-q4km:Q4_K_M
# Run inference directly in the terminal:
llama-cli -hf st4system/gemma4-e2b-scn-ollama-q4km: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 st4system/gemma4-e2b-scn-ollama-q4km:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf st4system/gemma4-e2b-scn-ollama-q4km: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 st4system/gemma4-e2b-scn-ollama-q4km:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf st4system/gemma4-e2b-scn-ollama-q4km:Q4_K_M

Use Docker

docker model run hf.co/st4system/gemma4-e2b-scn-ollama-q4km:Q4_K_M

LM Studio
Jan
Ollama
How to use st4system/gemma4-e2b-scn-ollama-q4km with Ollama:
```
ollama run hf.co/st4system/gemma4-e2b-scn-ollama-q4km:Q4_K_M
```

Unsloth Studio

How to use st4system/gemma4-e2b-scn-ollama-q4km 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 st4system/gemma4-e2b-scn-ollama-q4km 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 st4system/gemma4-e2b-scn-ollama-q4km to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for st4system/gemma4-e2b-scn-ollama-q4km to start chatting

Atomic Chat new
Docker Model Runner
How to use st4system/gemma4-e2b-scn-ollama-q4km with Docker Model Runner:
```
docker model run hf.co/st4system/gemma4-e2b-scn-ollama-q4km:Q4_K_M
```

Lemonade

How to use st4system/gemma4-e2b-scn-ollama-q4km with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull st4system/gemma4-e2b-scn-ollama-q4km:Q4_K_M

Run and chat with the model

lemonade run user.gemma4-e2b-scn-ollama-q4km-Q4_K_M

List all available models

lemonade list

You need to agree to share your contact information to access this model

This repository is publicly accessible, but you have to accept the conditions to access its files and content.

gemma4-e2b-scn-ollama-q4km

Direct Ollama distribution of the SCN Gemma4 artifact.

What this is

monolithic GGUF for Ollama local import
image + text capable
audio removed by design
derived from the official Ollama Gemma4 monolithic blob, then audio tensors / metadata were pruned
packaged so that hf download -> ollama create -f Modelfile is enough
Related Hugging Face checkpoint repo: st4system/gemma4-e2b-scn

Why this repo exists

The older split GGUF + mmproj path is appropriate for llama.cpp, but not for direct Ollama Gemma4 vision use. This repo contains the single-GGUF artifact required by Ollama users.

Measured effect

source Ollama GGUF blob size before pruning: 6.7G
GGUF file size: 6.1G
removed audio metadata fields: 6
removed audio tensors: 752
kept tensors after pruning: 1,260
general.parameter_count after pruning: 4,815,994,851
size delta after pruning: 0.6G saved (8.58%)
quantization: q4_k_m

Validated

ollama show reports vision
Ollama text runtime passed
Ollama image runtime passed on the sample image in this project
audio is unsupported by design
PNG input was validated for image runtime

Install for Ollama users

HF_HUB_DISABLE_XET=1 hf download st4system/gemma4-e2b-scn-ollama-q4km --repo-type model --local-dir ~/models/gemma4-e2b-scn-ollama-q4km
cd ~/models/gemma4-e2b-scn-ollama-q4km
ollama create gemma4-scn -f Modelfile
ollama run gemma4-scn "こんにちは"

Vision smoke:

ollama show gemma4-scn

If Capabilities includes vision, the model is packaged correctly.

For actual image inference, prefer PNG or JPEG inputs in Ollama clients.

Files

gemma4-e2b-scn-q4_k_m.gguf
Modelfile
ollama_manifest.json
SHA256SUMS
NOTICE.txt

Compliance notes

This is a derivative of google/gemma-4-e2b-it
This is not an official Google release
OpenAI / ChatGPT generated training data was not used in this project

Downloads last month: -

GGUF

Model size

5B params

Architecture

gemma4

Hardware compatibility

4-bit

Inference Providers NEW

This model isn't deployed by any Inference Provider. 🙋 Ask for provider support