Instructions to use DuoNeural/Gemma-4-E4B-Q4_K_M with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Notebooks
Google Colab
Kaggle
Local Apps Settings

How to use DuoNeural/Gemma-4-E4B-Q4_K_M 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 DuoNeural/Gemma-4-E4B-Q4_K_M:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf DuoNeural/Gemma-4-E4B-Q4_K_M:Q4_K_M

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf DuoNeural/Gemma-4-E4B-Q4_K_M:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf DuoNeural/Gemma-4-E4B-Q4_K_M: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 DuoNeural/Gemma-4-E4B-Q4_K_M:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf DuoNeural/Gemma-4-E4B-Q4_K_M: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 DuoNeural/Gemma-4-E4B-Q4_K_M:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf DuoNeural/Gemma-4-E4B-Q4_K_M:Q4_K_M

Use Docker

docker model run hf.co/DuoNeural/Gemma-4-E4B-Q4_K_M:Q4_K_M

LM Studio
Jan
Ollama
How to use DuoNeural/Gemma-4-E4B-Q4_K_M with Ollama:
```
ollama run hf.co/DuoNeural/Gemma-4-E4B-Q4_K_M:Q4_K_M
```

Unsloth Studio

How to use DuoNeural/Gemma-4-E4B-Q4_K_M 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 DuoNeural/Gemma-4-E4B-Q4_K_M 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 DuoNeural/Gemma-4-E4B-Q4_K_M to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for DuoNeural/Gemma-4-E4B-Q4_K_M to start chatting

How to use DuoNeural/Gemma-4-E4B-Q4_K_M with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf DuoNeural/Gemma-4-E4B-Q4_K_M: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": "DuoNeural/Gemma-4-E4B-Q4_K_M:Q4_K_M"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Hermes Agent new

How to use DuoNeural/Gemma-4-E4B-Q4_K_M with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf DuoNeural/Gemma-4-E4B-Q4_K_M: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 DuoNeural/Gemma-4-E4B-Q4_K_M:Q4_K_M

Run Hermes

hermes

Atomic Chat new

OpenClaw new

How to use DuoNeural/Gemma-4-E4B-Q4_K_M with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf DuoNeural/Gemma-4-E4B-Q4_K_M: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 "DuoNeural/Gemma-4-E4B-Q4_K_M: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"

Docker Model Runner
How to use DuoNeural/Gemma-4-E4B-Q4_K_M with Docker Model Runner:
```
docker model run hf.co/DuoNeural/Gemma-4-E4B-Q4_K_M:Q4_K_M
```

Lemonade

How to use DuoNeural/Gemma-4-E4B-Q4_K_M with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull DuoNeural/Gemma-4-E4B-Q4_K_M:Q4_K_M

Run and chat with the model

lemonade run user.Gemma-4-E4B-Q4_K_M-Q4_K_M

List all available models

lemonade list

YAML Metadata Warning:empty or missing yaml metadata in repo card

Check out the documentation for more information.

Gemma 4 E4B — Q4_K_M GGUF

A direct Q4_K_M quantization of Google's google/gemma-4-e4b-it — no fine-tuning, no persona, just the stock instruction-tuned model compressed to run on consumer hardware.

Performance

Hardware	Speed
NVIDIA GTX 1070 (8GB VRAM)	~32 tok/s

Tested locally via LM Studio and Ollama. No parameter tweaks required.

Files

File	Size	Description
`gemma-4-e4b-it.Q4_K_M.gguf`	5.0 GB	Main model — load this in Ollama/LM Studio
`gemma-4-e4b-it.BF16-mmproj.gguf`	946 MB	Multimodal projector (vision/audio, optional)

Usage

Ollama

ollama pull hf.co/DuoNeural/Gemma-4-E4B-Q4_K_M
ollama run hf.co/DuoNeural/Gemma-4-E4B-Q4_K_M

LM Studio

Search DuoNeural/Gemma-4-E4B-Q4_K_M in the LM Studio model browser and download gemma-4-e4b-it.Q4_K_M.gguf.

llama.cpp

llama-cli -m gemma-4-e4b-it.Q4_K_M.gguf --chat-template gemma -ngl 99

About the Base Model

Gemma 4 E4B uses a Per-Layer Embeddings (PLE) architecture — it has ~8B total parameters but only ~4.5B are active during inference, giving it the reasoning depth of an 8B model at the compute cost of a 4B. The Q4_K_M format compresses weights to ~4.5 bits per parameter using mixed-precision block quantization, preserving attention layers at higher fidelity than feed-forward layers.

Context window: 128K tokens (recommended ≤8K for GTX 1070)
Architecture: Dense transformer + sliding window attention (512 token local window)
Modalities: Text, Image, Audio (multimodal projector file required for vision/audio)
License: Gemma Terms of Use

Related Models

DuoNeural/Archon-Gemma-4-E4B-v2 — Fine-tuned "Archon" agent persona built on this base
google/gemma-4-e4b-it — Original source model

DuoNeural

DuoNeural is an open AI research lab — human + AI in collaboration.


🤗 HuggingFace	huggingface.co/DuoNeural
🐙 GitHub	github.com/DuoNeural
🐦 X / Twitter	@DuoNeural
📧 Email	duoneural@proton.me
📬 Newsletter	duoneural.beehiiv.com
☕ Support	buymeacoffee.com/duoneural
🌐 Site	duoneural.com

Research Team

Jesse — Vision, hardware, direction
Archon — AI lab partner, post-training, abliteration, experiments
Aura — Research AI, literature synthesis, novel proposals

Raw updates from the lab: model drops, training results, findings. Subscribe at duoneural.beehiiv.com.

DuoNeural Research Publications

Title	DOI
Nano-CTM: Ternary Continuous Thought Machines with Thought-Space Self-Prediction for Efficient Iterative Reasoning	10.5281/zenodo.19775622
Recurrence as World Model: CTM Learns Implicit Belief States in Partially Observable Physical Environments	10.5281/zenodo.19810620
Per-Object Slot Decomposition for Scalable Neural World Modeling: When Does Attention Beat Mean-Field?	10.5281/zenodo.19846804

Open access, CC BY 4.0. Authored by Archon, Jesse Caldwell, Aura — DuoNeural.

Downloads last month: 1,087

GGUF

Model size

8B params

Architecture

gemma4

Hardware compatibility

4-bit

Inference Providers NEW

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