Instructions to use deucebucket/Gemma-4-26B-A4B-it-Cerebellum-v6-GGUF with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Notebooks
Google Colab
Kaggle
Local Apps Settings

How to use deucebucket/Gemma-4-26B-A4B-it-Cerebellum-v6-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 deucebucket/Gemma-4-26B-A4B-it-Cerebellum-v6-GGUF:Q3_K_M
# Run inference directly in the terminal:
llama cli -hf deucebucket/Gemma-4-26B-A4B-it-Cerebellum-v6-GGUF:Q3_K_M

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf deucebucket/Gemma-4-26B-A4B-it-Cerebellum-v6-GGUF:Q3_K_M
# Run inference directly in the terminal:
llama cli -hf deucebucket/Gemma-4-26B-A4B-it-Cerebellum-v6-GGUF:Q3_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 deucebucket/Gemma-4-26B-A4B-it-Cerebellum-v6-GGUF:Q3_K_M
# Run inference directly in the terminal:
./llama-cli -hf deucebucket/Gemma-4-26B-A4B-it-Cerebellum-v6-GGUF:Q3_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 deucebucket/Gemma-4-26B-A4B-it-Cerebellum-v6-GGUF:Q3_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf deucebucket/Gemma-4-26B-A4B-it-Cerebellum-v6-GGUF:Q3_K_M

Use Docker

docker model run hf.co/deucebucket/Gemma-4-26B-A4B-it-Cerebellum-v6-GGUF:Q3_K_M

LM Studio
Jan

vLLM

How to use deucebucket/Gemma-4-26B-A4B-it-Cerebellum-v6-GGUF with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "deucebucket/Gemma-4-26B-A4B-it-Cerebellum-v6-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": "deucebucket/Gemma-4-26B-A4B-it-Cerebellum-v6-GGUF",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/deucebucket/Gemma-4-26B-A4B-it-Cerebellum-v6-GGUF:Q3_K_M

Ollama
How to use deucebucket/Gemma-4-26B-A4B-it-Cerebellum-v6-GGUF with Ollama:
```
ollama run hf.co/deucebucket/Gemma-4-26B-A4B-it-Cerebellum-v6-GGUF:Q3_K_M
```
Unsloth Desktop

How to use deucebucket/Gemma-4-26B-A4B-it-Cerebellum-v6-GGUF with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf deucebucket/Gemma-4-26B-A4B-it-Cerebellum-v6-GGUF:Q3_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": "deucebucket/Gemma-4-26B-A4B-it-Cerebellum-v6-GGUF:Q3_K_M"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Docker Model Runner
How to use deucebucket/Gemma-4-26B-A4B-it-Cerebellum-v6-GGUF with Docker Model Runner:
```
docker model run hf.co/deucebucket/Gemma-4-26B-A4B-it-Cerebellum-v6-GGUF:Q3_K_M
```

Lemonade

How to use deucebucket/Gemma-4-26B-A4B-it-Cerebellum-v6-GGUF with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull deucebucket/Gemma-4-26B-A4B-it-Cerebellum-v6-GGUF:Q3_K_M

Run and chat with the model

lemonade run user.Gemma-4-26B-A4B-it-Cerebellum-v6-GGUF-Q3_K_M

List all available models

lemonade list

Hermes Agent

How to use deucebucket/Gemma-4-26B-A4B-it-Cerebellum-v6-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 deucebucket/Gemma-4-26B-A4B-it-Cerebellum-v6-GGUF:Q3_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 deucebucket/Gemma-4-26B-A4B-it-Cerebellum-v6-GGUF:Q3_K_M

Run Hermes

hermes

Atomic Chat

OpenClaw

How to use deucebucket/Gemma-4-26B-A4B-it-Cerebellum-v6-GGUF with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf deucebucket/Gemma-4-26B-A4B-it-Cerebellum-v6-GGUF:Q3_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 "deucebucket/Gemma-4-26B-A4B-it-Cerebellum-v6-GGUF:Q3_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-E4B-it-Cerebellum-v1

by dont-remember-it - opened May 2

Discussion

dont-remember-it

May 2

•

edited May 2

Hello,
I read about your Gemma-4-E4B-it-Cerebellum-v1 and v2 and was planning to try it out today. However, those models are no longer available. Are you planning to upload it again?
Thanks for your hard work on these models. I am looking forward to testing it out.

deucebucket

Owner May 2

Hey sorry, was moving things around, and accidentally privated it. Got v2 though! Let me know how it goes. These all have worked fine for me, but I'm just one guy testing what I can. deucebucket/Gemma-4-E4B-it-Cerebellum-v2-GGUF

dont-remember-it

May 2

Thank! Will try it out and let you know if I discover something substantial.

dont-remember-it

May 14

Hello, I just wanted to say Cerebellum v6 is genuinely one of the most impressive quantization efforts I've seen and tested. On my RTX 4070 Laptop (8GB VRAM) with the AtomicBot TurboQuant fork and MTP speculative decoding, Cerebellum took me from 2.4 tok/s to 30+ tok/s compared to a standard Q4_K_M, a 12x improvement! The response quality jumped noticeably too.

Do you have any plans to incorporate vision support in future version as base gemma model is natively multimodal?

Just flagging it as something the community would love. Thanks for putting in the work.

deucebucket

Owner May 15

Hello, I just wanted to say Cerebellum v6 is genuinely one of the most impressive quantization efforts I've seen and tested. On my RTX 4070 Laptop (8GB VRAM) with the AtomicBot TurboQuant fork and MTP speculative decoding, Cerebellum took me from 2.4 tok/s to 30+ tok/s compared to a standard Q4_K_M, a 12x improvement! The response quality jumped noticeably too.

Do you have any plans to incorporate vision support in future version as base gemma model is natively multimodal?

Just flagging it as something the community would love. Thanks for putting in the work.

yeah, its definitely on the board, my first runs targeted wikitext2 for the improvements. So I'm sure by that line of thinking, the same forward and backward ablation but geared towards getting action when given images, im sure it could be tooled to improve that, it would most likely increase the size, just not sure how much. No timeline as of yet, I'm still learning a lot about this, but definitely eventually! Thanks for trying the model!

deucebucket

Owner May 15

Hello, I just wanted to say Cerebellum v6 is genuinely one of the most impressive quantization efforts I've seen and tested. On my RTX 4070 Laptop (8GB VRAM) with the AtomicBot TurboQuant fork and MTP speculative decoding, Cerebellum took me from 2.4 tok/s to 30+ tok/s compared to a standard Q4_K_M, a 12x improvement! The response quality jumped noticeably too.

Do you have any plans to incorporate vision support in future version as base gemma model is natively multimodal?

Just flagging it as something the community would love. Thanks for putting in the work.

so it was a lot simpler, I didnt include the mmproj gguf required for vision, this was an oversite, i include the text backbone, but never shipped the vision gguf, should be up on the repo soon, or bartowkis would work also.

dont-remember-it

May 15

Just tested it and it is working seamlessly. Thanks once again!

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