Instructions to use mradermacher/gemma-4-E4B-it-OBLITERATED-i1-GGUF with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use mradermacher/gemma-4-E4B-it-OBLITERATED-i1-GGUF with Transformers:

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

pipe = pipeline("text-generation", model="mradermacher/gemma-4-E4B-it-OBLITERATED-i1-GGUF")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoModel
model = AutoModel.from_pretrained("mradermacher/gemma-4-E4B-it-OBLITERATED-i1-GGUF", dtype="auto")

llama-cpp-python

How to use mradermacher/gemma-4-E4B-it-OBLITERATED-i1-GGUF with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="mradermacher/gemma-4-E4B-it-OBLITERATED-i1-GGUF",
	filename="gemma-4-E4B-it-OBLITERATED.i1-IQ1_M.gguf",
)

llm.create_chat_completion(
	messages = [
		{
			"role": "user",
			"content": "What is the capital of France?"
		}
	]
)

Notebooks
Google Colab
Kaggle
Local Apps Settings

llama.cpp

How to use mradermacher/gemma-4-E4B-it-OBLITERATED-i1-GGUF with llama.cpp:

Install from brew

brew install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf mradermacher/gemma-4-E4B-it-OBLITERATED-i1-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama-cli -hf mradermacher/gemma-4-E4B-it-OBLITERATED-i1-GGUF:Q4_K_M

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf mradermacher/gemma-4-E4B-it-OBLITERATED-i1-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama-cli -hf mradermacher/gemma-4-E4B-it-OBLITERATED-i1-GGUF: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 mradermacher/gemma-4-E4B-it-OBLITERATED-i1-GGUF:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf mradermacher/gemma-4-E4B-it-OBLITERATED-i1-GGUF: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 mradermacher/gemma-4-E4B-it-OBLITERATED-i1-GGUF:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf mradermacher/gemma-4-E4B-it-OBLITERATED-i1-GGUF:Q4_K_M

Use Docker

docker model run hf.co/mradermacher/gemma-4-E4B-it-OBLITERATED-i1-GGUF:Q4_K_M

LM Studio
Jan

vLLM

How to use mradermacher/gemma-4-E4B-it-OBLITERATED-i1-GGUF with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "mradermacher/gemma-4-E4B-it-OBLITERATED-i1-GGUF"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "mradermacher/gemma-4-E4B-it-OBLITERATED-i1-GGUF",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/mradermacher/gemma-4-E4B-it-OBLITERATED-i1-GGUF:Q4_K_M

SGLang

How to use mradermacher/gemma-4-E4B-it-OBLITERATED-i1-GGUF 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 "mradermacher/gemma-4-E4B-it-OBLITERATED-i1-GGUF" \
    --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": "mradermacher/gemma-4-E4B-it-OBLITERATED-i1-GGUF",
		"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 "mradermacher/gemma-4-E4B-it-OBLITERATED-i1-GGUF" \
        --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": "mradermacher/gemma-4-E4B-it-OBLITERATED-i1-GGUF",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Ollama
How to use mradermacher/gemma-4-E4B-it-OBLITERATED-i1-GGUF with Ollama:
```
ollama run hf.co/mradermacher/gemma-4-E4B-it-OBLITERATED-i1-GGUF:Q4_K_M
```

Unsloth Studio

How to use mradermacher/gemma-4-E4B-it-OBLITERATED-i1-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 mradermacher/gemma-4-E4B-it-OBLITERATED-i1-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 mradermacher/gemma-4-E4B-it-OBLITERATED-i1-GGUF to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for mradermacher/gemma-4-E4B-it-OBLITERATED-i1-GGUF to start chatting

How to use mradermacher/gemma-4-E4B-it-OBLITERATED-i1-GGUF with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama-server -hf mradermacher/gemma-4-E4B-it-OBLITERATED-i1-GGUF:Q4_K_M

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": "mradermacher/gemma-4-E4B-it-OBLITERATED-i1-GGUF:Q4_K_M"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Hermes Agent new

How to use mradermacher/gemma-4-E4B-it-OBLITERATED-i1-GGUF with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama-server -hf mradermacher/gemma-4-E4B-it-OBLITERATED-i1-GGUF: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 mradermacher/gemma-4-E4B-it-OBLITERATED-i1-GGUF:Q4_K_M

Run Hermes

hermes

Docker Model Runner
How to use mradermacher/gemma-4-E4B-it-OBLITERATED-i1-GGUF with Docker Model Runner:
```
docker model run hf.co/mradermacher/gemma-4-E4B-it-OBLITERATED-i1-GGUF:Q4_K_M
```

Lemonade

How to use mradermacher/gemma-4-E4B-it-OBLITERATED-i1-GGUF with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull mradermacher/gemma-4-E4B-it-OBLITERATED-i1-GGUF:Q4_K_M

Run and chat with the model

lemonade run user.gemma-4-E4B-it-OBLITERATED-i1-GGUF-Q4_K_M

List all available models

lemonade list

can't load the model for some reason

by klam97xd426 - opened Apr 17

Discussion

klam97xd426

Apr 17

Can't load your GGUFs of this model in LM Studio for some reason. I keep getting missing tensor 'blk.24.attn_k.weight' . Tried both 5_K_M and 4_K_M of both the static and imatrix quants.

I've been using the same setting with other E4B GGUFs that you made and they load fine... Thanks again for your work. Really appreciate it as always.

RichardErkhov

Apr 17

hi, try updating your llama cpp, perhaps it's too outdated. Latest llama cpp works:

llama-cli.exe --model gemma-4-E4B-it-OBLITERATED.i1-Q5_K_M.gguf --reasoning on
load_backend: loaded RPC backend from C:\Users\Administrator\Downloads\llama-b8827-bin-win-cuda-13.1-x64\ggml-rpc.dll
load_backend: loaded CPU backend from C:\Users\Administrator\Downloads\llama-b8827-bin-win-cuda-13.1-x64\ggml-cpu-alderlake.dll

Loading model...


▄▄ ▄▄
██ ██
██ ██  ▀▀█▄ ███▄███▄  ▀▀█▄    ▄████ ████▄ ████▄
██ ██ ▄█▀██ ██ ██ ██ ▄█▀██    ██    ██ ██ ██ ██
██ ██ ▀█▄██ ██ ██ ██ ▀█▄██ ██ ▀████ ████▀ ████▀
                                    ██    ██
                                    ▀▀    ▀▀

build      : b8827-5e6c0e18b
model      : gemma-4-E4B-it-OBLITERATED.i1-Q5_K_M.gguf
modalities : text

available commands:
  /exit or Ctrl+C     stop or exit
  /regen              regenerate the last response
  /clear              clear the chat history
  /read <file>        add a text file
  /glob <pattern>     add text files using globbing pattern


> hey

[Start thinking]
1.  **Analyze the user's input:** The user said "hey." It's a very casual, generic greeting.
2.  **Determine the desired response type:** The response should be friendly, welcoming, and conversational but doesn't need to be huge, allowing the user to easily take the next turn.
3.  **Brainstorm options:**
    *   "Hey!" (Too short, maybe not enough personality)
    *   "What's up?" (Very casual, good)
    *   "Hello there. How can I help you?" (Polite, slightly formal)
    *   "Hi! What's on your mind?") (Friendly, inviting)
4.  **Select the best option:** "Hey! How can I help you today?" is a good all-rounder—it matches the tone, is easygoing, and tells the user that I'm available to chat/be helpful).

[End thinking]

Hey! What's on your mind?

[ Prompt: 24.1 t/s | Generation: 4.1 t/s ]

klam97xd426

Apr 17

That's probably the reason why. The llama.cpp in LM Studio for linux's app image is still on

'llama.cpp release b8733 (commit d6f3030)'

I'll try to load it manually when I get home. Thanks for the help.

klam97xd426 changed discussion status to closed Apr 17

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