Instructions to use unsloth/gemma-4-E4B-it-qat-GGUF with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use unsloth/gemma-4-E4B-it-qat-GGUF with Transformers:

# Load model directly
from transformers import AutoProcessor, AutoModelForMultimodalLM

processor = AutoProcessor.from_pretrained("unsloth/gemma-4-E4B-it-qat-GGUF")
model = AutoModelForMultimodalLM.from_pretrained("unsloth/gemma-4-E4B-it-qat-GGUF", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

llama.cpp

How to use unsloth/gemma-4-E4B-it-qat-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 unsloth/gemma-4-E4B-it-qat-GGUF:UD-Q4_K_XL
# Run inference directly in the terminal:
llama cli -hf unsloth/gemma-4-E4B-it-qat-GGUF:UD-Q4_K_XL

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf unsloth/gemma-4-E4B-it-qat-GGUF:UD-Q4_K_XL
# Run inference directly in the terminal:
llama cli -hf unsloth/gemma-4-E4B-it-qat-GGUF:UD-Q4_K_XL

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 unsloth/gemma-4-E4B-it-qat-GGUF:UD-Q4_K_XL
# Run inference directly in the terminal:
./llama-cli -hf unsloth/gemma-4-E4B-it-qat-GGUF:UD-Q4_K_XL

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 unsloth/gemma-4-E4B-it-qat-GGUF:UD-Q4_K_XL
# Run inference directly in the terminal:
./build/bin/llama-cli -hf unsloth/gemma-4-E4B-it-qat-GGUF:UD-Q4_K_XL

Use Docker

docker model run hf.co/unsloth/gemma-4-E4B-it-qat-GGUF:UD-Q4_K_XL

LM Studio
Jan
Ollama
How to use unsloth/gemma-4-E4B-it-qat-GGUF with Ollama:
```
ollama run hf.co/unsloth/gemma-4-E4B-it-qat-GGUF:UD-Q4_K_XL
```

Unsloth Studio

How to use unsloth/gemma-4-E4B-it-qat-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 unsloth/gemma-4-E4B-it-qat-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 unsloth/gemma-4-E4B-it-qat-GGUF to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for unsloth/gemma-4-E4B-it-qat-GGUF to start chatting

How to use unsloth/gemma-4-E4B-it-qat-GGUF with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf unsloth/gemma-4-E4B-it-qat-GGUF:UD-Q4_K_XL

Configure the model in Pi

# Install Pi:
npm install -g @mariozechner/pi-coding-agent
# Add to ~/.pi/agent/models.json:
{
  "providers": {
    "llama-cpp": {
      "baseUrl": "http://localhost:8080/v1",
      "api": "openai-completions",
      "apiKey": "none",
      "models": [
        {
          "id": "unsloth/gemma-4-E4B-it-qat-GGUF:UD-Q4_K_XL"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

OpenClaw new

How to use unsloth/gemma-4-E4B-it-qat-GGUF with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf unsloth/gemma-4-E4B-it-qat-GGUF:UD-Q4_K_XL

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 "unsloth/gemma-4-E4B-it-qat-GGUF:UD-Q4_K_XL" \
  --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"

Docker Model Runner
How to use unsloth/gemma-4-E4B-it-qat-GGUF with Docker Model Runner:
```
docker model run hf.co/unsloth/gemma-4-E4B-it-qat-GGUF:UD-Q4_K_XL
```

Lemonade

How to use unsloth/gemma-4-E4B-it-qat-GGUF with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull unsloth/gemma-4-E4B-it-qat-GGUF:UD-Q4_K_XL

Run and chat with the model

lemonade run user.gemma-4-E4B-it-qat-GGUF-UD-Q4_K_XL

List all available models

lemonade list

Hermes Agent

How to use unsloth/gemma-4-E4B-it-qat-GGUF with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf unsloth/gemma-4-E4B-it-qat-GGUF:UD-Q4_K_XL

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 unsloth/gemma-4-E4B-it-qat-GGUF:UD-Q4_K_XL

Run Hermes

hermes

Atomic Chat

Performance report: 55 t/s on RTX 4060

by SlavikF - opened Jul 8

Discussion

SlavikF

Jul 8

I downloaded the model:

curl -LsSf https://hf.co/cli/install.sh | bash
sudo /home/slavik/.local/bin/hf download unsloth/gemma-4-E4B-it-qat-GGUF --include *UD-Q4_K_XL.gguf --include *-BF16.gguf

On Nvidia RTX 4060 with 8GB VRAM, I'm getting:

Prompt processing: 2500 t/s
Token generation: 55 t/s
VRAM used: 6.2 Gi

running it on k3s on Debian:

apiVersion: apps/v1
kind: Deployment
metadata:
  name: ai-small
  labels:
    app.kubernetes.io/name: ai-small
spec:
  selector:
    matchLabels:
      app.kubernetes.io/name: ai-small
  strategy:
    type: Recreate
  template:
    metadata:
      labels:
        app.kubernetes.io/name: ai-small
    spec:
      runtimeClassName: nvidia
      containers:
        - name: ai-small
          image: ghcr.io/ggml-org/llama.cpp:server-cuda12-b9776
          imagePullPolicy: IfNotPresent
          resources:
            requests:
              cpu: 20m
              memory: "4Gi"
            limits:
              memory: "10Gi"
              cpu: 2
          env:
            - name: NVIDIA_VISIBLE_DEVICES
              value: all
            - name: NVIDIA_DRIVER_CAPABILITIES
              value: all
          command: ["./llama-server"]
          args:
            - "--host"
            - "0.0.0.0"
            - "--port"
            - "30060"
            - "--cache-ram"
            - "4096"
            - "--tools"
            - "all"
            - "--models-max"
            - "1"
            - "--models-preset"
            - "/app/models.ini"
          volumeMounts:
            - mountPath: /root/.cache
              name: host-cache
            - mountPath: /app/models.ini
              name: models
              subPath: models.ini
      volumes:
        - name: host-cache
          hostPath:
            path: /root/.cache
            type: DirectoryOrCreate
        - name: models
          configMap:
            name: ai-small-models
---
apiVersion: v1
kind: Service
metadata:
  name: ai-small
  labels:
    app.kubernetes.io/name: ai-small
spec:
  type: NodePort
  ports:
    - name: http
      port: 30060
      targetPort: 30060
      nodePort: 30060
  selector:
    app.kubernetes.io/name: ai-small
---
apiVersion: v1
kind: ConfigMap
metadata:
  name: ai-small-models
  labels:
    app.kubernetes.io/name: ai-small
data:
  models.ini: |
    version = 1

    [unsloth/gemma-4-E4B-it-qat-GGUF:Q4_K_XL]
    alias=local-vla-gemma4B
    ctx-size=131072
    fit=off
    gpu-layers=all
    load-on-startup=true

Kwissbeats

about 1 month ago

that is some decent Prompt processing, but I imagined Token generation to be faster on a 4060? I only have a 5060 and haven't tried this yet.

Upload images, audio, and videos by dragging in the text input, pasting, or clicking here.

Tap or paste here to upload images

· Sign up or log in to comment