Instructions to use sarrington/gemma-4-e2b-spliced with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Notebooks
Google Colab
Kaggle
Local Apps Settings

How to use sarrington/gemma-4-e2b-spliced 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 sarrington/gemma-4-e2b-spliced:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf sarrington/gemma-4-e2b-spliced:Q4_K_M

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf sarrington/gemma-4-e2b-spliced:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf sarrington/gemma-4-e2b-spliced: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 sarrington/gemma-4-e2b-spliced:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf sarrington/gemma-4-e2b-spliced: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 sarrington/gemma-4-e2b-spliced:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf sarrington/gemma-4-e2b-spliced:Q4_K_M

Use Docker

docker model run hf.co/sarrington/gemma-4-e2b-spliced:Q4_K_M

LM Studio
Jan

vLLM

How to use sarrington/gemma-4-e2b-spliced with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "sarrington/gemma-4-e2b-spliced"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "sarrington/gemma-4-e2b-spliced",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/sarrington/gemma-4-e2b-spliced:Q4_K_M

Ollama
How to use sarrington/gemma-4-e2b-spliced with Ollama:
```
ollama run hf.co/sarrington/gemma-4-e2b-spliced:Q4_K_M
```
Unsloth Desktop

How to use sarrington/gemma-4-e2b-spliced with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf sarrington/gemma-4-e2b-spliced:Q4_K_M

Configure the model in Pi

# Install Pi:
npm install -g @earendil-works/pi-coding-agent
# Add to ~/.pi/agent/models.json:
{
  "providers": {
    "llama-cpp": {
      "baseUrl": "http://localhost:8080/v1",
      "api": "openai-completions",
      "apiKey": "none",
      "models": [
        {
          "id": "sarrington/gemma-4-e2b-spliced:Q4_K_M"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Docker Model Runner
How to use sarrington/gemma-4-e2b-spliced with Docker Model Runner:
```
docker model run hf.co/sarrington/gemma-4-e2b-spliced:Q4_K_M
```

Lemonade

How to use sarrington/gemma-4-e2b-spliced with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull sarrington/gemma-4-e2b-spliced:Q4_K_M

Run and chat with the model

lemonade run user.gemma-4-e2b-spliced-Q4_K_M

List all available models

lemonade list

Hermes Agent

How to use sarrington/gemma-4-e2b-spliced with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf sarrington/gemma-4-e2b-spliced:Q4_K_M

Configure Hermes

# Install Hermes:
curl -fsSL https://hermes-agent.nousresearch.com/install.sh | bash
hermes setup
# Point Hermes at the local server:
hermes config set model.provider custom
hermes config set model.base_url http://127.0.0.1:8080/v1
hermes config set model.default sarrington/gemma-4-e2b-spliced:Q4_K_M

Run Hermes

hermes

Atomic Chat

OpenClaw

How to use sarrington/gemma-4-e2b-spliced with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf sarrington/gemma-4-e2b-spliced:Q4_K_M

Configure OpenClaw

# Install OpenClaw:
npm install -g openclaw@latest
# Register the local server and set it as the default model:
openclaw onboard --non-interactive --mode local \
  --auth-choice custom-api-key \
  --custom-base-url http://127.0.0.1:8080/v1 \
  --custom-model-id "sarrington/gemma-4-e2b-spliced:Q4_K_M" \
  --custom-provider-id llama-cpp \
  --custom-compatibility openai \
  --custom-text-input \
  --accept-risk \
  --skip-health

Run OpenClaw

openclaw agent --local --agent main --message "Hello from Hugging Face"

Gemma 4 Edge Spliced GGUF

A highly compressed, spliced variant of Google's lightweight Gemma 4 E2B-it (Effective 2 Billion, also known as E4B with 4.65 Billion active parameters) model.

Model Description

This repository contains a spliced mixed-precision GGUF binary of the Gemma 4 E2B-it architecture. By surgically removing redundant layers, this model is optimized for hyper-fast, whisper-silent execution on low-memory edge platforms and commodity Apple Silicon hardware.

Format: GGUF (Q4_K_M)
Size: 3.45 GB
Target Platforms: Apple Silicon MacBooks (M1/M2/M3/M4) and standard CPU/GPU local runtimes.

Local Quickstart

Run the model natively via standard llama-cli:

llama-cli -ngl 100 -m ./google_gemma-4-E2B-it-Q4_K_M.gguf -p "The beauty of edge-first AI lies in" -n 128

Downloads last month: 5

GGUF

Model size

5B params

Architecture

gemma4

Hardware compatibility

4-bit