Instructions to use 888rok/gemma-4-E2B-it-wllama-split with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Notebooks
Google Colab
Kaggle
Local Apps Settings

How to use 888rok/gemma-4-E2B-it-wllama-split 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 888rok/gemma-4-E2B-it-wllama-split:UD-IQ2_M
# Run inference directly in the terminal:
llama cli -hf 888rok/gemma-4-E2B-it-wllama-split:UD-IQ2_M

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf 888rok/gemma-4-E2B-it-wllama-split:UD-IQ2_M
# Run inference directly in the terminal:
llama cli -hf 888rok/gemma-4-E2B-it-wllama-split:UD-IQ2_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 888rok/gemma-4-E2B-it-wllama-split:UD-IQ2_M
# Run inference directly in the terminal:
./llama-cli -hf 888rok/gemma-4-E2B-it-wllama-split:UD-IQ2_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 888rok/gemma-4-E2B-it-wllama-split:UD-IQ2_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf 888rok/gemma-4-E2B-it-wllama-split:UD-IQ2_M

Use Docker

docker model run hf.co/888rok/gemma-4-E2B-it-wllama-split:UD-IQ2_M

LM Studio
Jan
Ollama
How to use 888rok/gemma-4-E2B-it-wllama-split with Ollama:
```
ollama run hf.co/888rok/gemma-4-E2B-it-wllama-split:UD-IQ2_M
```
Unsloth Desktop

How to use 888rok/gemma-4-E2B-it-wllama-split with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf 888rok/gemma-4-E2B-it-wllama-split:UD-IQ2_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": "888rok/gemma-4-E2B-it-wllama-split:UD-IQ2_M"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Docker Model Runner
How to use 888rok/gemma-4-E2B-it-wllama-split with Docker Model Runner:
```
docker model run hf.co/888rok/gemma-4-E2B-it-wllama-split:UD-IQ2_M
```

Lemonade

How to use 888rok/gemma-4-E2B-it-wllama-split with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull 888rok/gemma-4-E2B-it-wllama-split:UD-IQ2_M

Run and chat with the model

lemonade run user.gemma-4-E2B-it-wllama-split-UD-IQ2_M

List all available models

lemonade list

Hermes Agent

How to use 888rok/gemma-4-E2B-it-wllama-split with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf 888rok/gemma-4-E2B-it-wllama-split:UD-IQ2_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 888rok/gemma-4-E2B-it-wllama-split:UD-IQ2_M

Run Hermes

hermes

Atomic Chat

OpenClaw

How to use 888rok/gemma-4-E2B-it-wllama-split with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf 888rok/gemma-4-E2B-it-wllama-split:UD-IQ2_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 "888rok/gemma-4-E2B-it-wllama-split:UD-IQ2_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 E2B IT — UD-IQ2_M, split for wllama

This is unsloth/gemma-4-E2B-it-GGUF's gemma-4-E2B-it-UD-IQ2_M.gguf (2.29 GB), split into 3 shards with llama-gguf-split --split-max-size 1G so it can run in the browser with wllama, which cannot load single files over 2 GB (ArrayBuffer limit).

Load it by pointing wllama at the first shard — the rest resolve automatically:

https://huggingface.co/888rok/gemma-4-E2B-it-wllama-split/resolve/main/gemma-4-E2B-it-UD-IQ2_M-00001-of-00003.gguf

All credit for the model to Google (Gemma 4) and for the quantization to Unsloth.

Downloads last month: 31

GGUF

Model size

5B params

Architecture

gemma4

Hardware compatibility

2-bit

Inference Providers NEW

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

Model tree for 888rok/gemma-4-E2B-it-wllama-split

Base model

google/gemma-4-E2B

Finetuned

google/gemma-4-E2B-it

Quantized

unsloth/gemma-4-E2B-it-GGUF

Quantized

(3)

this model