Instructions to use sphaela/gemma-4-E2B-it-AutoRound-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 sphaela/gemma-4-E2B-it-AutoRound-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 sphaela/gemma-4-E2B-it-AutoRound-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf sphaela/gemma-4-E2B-it-AutoRound-GGUF:Q4_K_M

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf sphaela/gemma-4-E2B-it-AutoRound-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf sphaela/gemma-4-E2B-it-AutoRound-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 sphaela/gemma-4-E2B-it-AutoRound-GGUF:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf sphaela/gemma-4-E2B-it-AutoRound-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 sphaela/gemma-4-E2B-it-AutoRound-GGUF:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf sphaela/gemma-4-E2B-it-AutoRound-GGUF:Q4_K_M

Use Docker

docker model run hf.co/sphaela/gemma-4-E2B-it-AutoRound-GGUF:Q4_K_M

LM Studio
Jan
Ollama
How to use sphaela/gemma-4-E2B-it-AutoRound-GGUF with Ollama:
```
ollama run hf.co/sphaela/gemma-4-E2B-it-AutoRound-GGUF:Q4_K_M
```
Unsloth Desktop

How to use sphaela/gemma-4-E2B-it-AutoRound-GGUF with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf sphaela/gemma-4-E2B-it-AutoRound-GGUF:Q4_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": "sphaela/gemma-4-E2B-it-AutoRound-GGUF:Q4_K_M"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Docker Model Runner
How to use sphaela/gemma-4-E2B-it-AutoRound-GGUF with Docker Model Runner:
```
docker model run hf.co/sphaela/gemma-4-E2B-it-AutoRound-GGUF:Q4_K_M
```

Lemonade

How to use sphaela/gemma-4-E2B-it-AutoRound-GGUF with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull sphaela/gemma-4-E2B-it-AutoRound-GGUF:Q4_K_M

Run and chat with the model

lemonade run user.gemma-4-E2B-it-AutoRound-GGUF-Q4_K_M

List all available models

lemonade list

Hermes Agent

How to use sphaela/gemma-4-E2B-it-AutoRound-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 sphaela/gemma-4-E2B-it-AutoRound-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 sphaela/gemma-4-E2B-it-AutoRound-GGUF:Q4_K_M

Run Hermes

hermes

Atomic Chat

OpenClaw

How to use sphaela/gemma-4-E2B-it-AutoRound-GGUF with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf sphaela/gemma-4-E2B-it-AutoRound-GGUF: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 "sphaela/gemma-4-E2B-it-AutoRound-GGUF: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"

gemma-4-E2B-it-AutoRound-GGUF / README.md

sphaela

Update README.md

e89cab1 verified 6 months ago

preview code

Raw History Blame

3.42 kB

metadata

license: apache-2.0
language:
  - multilingual
base_model: google/gemma-4-E2B-it
tags:
  - auto-round
  - intel
  - gguf
  - quantization
  - vlm

Gemma-4-E2B-it GGUF (AutoRound Quantized)

This repository contains GGUF quantized versions of google/gemma-4-E2B-it created using Intel's AutoRound quantization method.

Quantization Details

The models were quantized using various schemes provided by the auto-round tool. For better compatibility and smaller size, we provide unified multimodal projector (mmproj) files in F16, BF16, and F32 formats.

Files and Sizes

File Name	Quant Type	Size	Description
`gemma-4-E2B-it-Q2_K_S.gguf`	Q2_K_S	2.7 GB	Extremely high compression, significant quality loss.
`gemma-4-E2B-it-Q2_K_MIXED.gguf`	Q2_K_MIXED	3.8 GB	Recommended high-compression option. Uses Q4 for KV cache with good quality.
`gemma-4-E2B-it-Q3_K_S.gguf`	Q3_K_S	2.9 GB	Very high compression, notable quality loss.
`gemma-4-E2B-it-Q3_K_M.gguf`	Q3_K_M	2.9 GB	Balanced 3-bit quantization.
`gemma-4-E2B-it-Q3_K_L.gguf`	Q3_K_L	2.9 GB	High quality 3-bit quantization.
`gemma-4-E2B-it-Q4_0.gguf`	Q4_0	3.2 GB	Standard 4-bit quantization, good balance.
`gemma-4-E2B-it-Q4_1.gguf`	Q4_1	3.3 GB	Higher quality 4-bit quantization than Q4_0.
`gemma-4-E2B-it-Q4_K_S.gguf`	Q4_K_S	3.2 GB	Small 4-bit K-quant, good efficiency.
`gemma-4-E2B-it-Q4_K_M.gguf`	Q4_K_M	3.2 GB	Recommended 4-bit K-quant, excellent balance.
`gemma-4-E2B-it-Q5_0.gguf`	Q5_0	3.4 GB	Standard 5-bit quantization, very high quality.
`gemma-4-E2B-it-Q5_1.gguf`	Q5_1	3.5 GB	Higher quality 5-bit quantization than Q5_0.
`gemma-4-E2B-it-Q5_K_S.gguf`	Q5_K_S	3.4 GB	Small 5-bit K-quant, very high quality.
`gemma-4-E2B-it-Q5_K_M.gguf`	Q5_K_M	3.4 GB	Recommended 5-bit K-quant, near-lossless.
`gemma-4-E2B-it-Q6_K.gguf`	Q6_K	3.6 GB	6-bit K-quant, virtually indistinguishable from F16.
`gemma-4-E2B-it-Q8_0.gguf`	Q8_0	4.7 GB	8-bit quantization, near-lossless.
`mmproj-model-f16.gguf`	F16	940 MB	Unified Projector in Float16 format.
`mmproj-model-bf16.gguf`	BF16	942 MB	Unified Projector in BFloat16 format.
`mmproj-model-f32.gguf`	F32	1.8 GB	Unified Projector in Float32 format.

Generate the Model

The models were generated using Intel's AutoRound with the following command:

auto-round --model google/gemma-4-E2B-it --output_dir ./quantized/ --scheme <SCHEME> --iters 0

Note: To reproduce this quantization, you need the following Pull Request:

Intel AutoRound PR #1655
And the specific fix mentioned in this comment.

Usage with llama.cpp

These models can be used with llama.cpp. For multimodal usage, you must specify the projector file:

./llama-cli -m gemma-4-E2B-it-Q4_K_M.gguf --mmproj mmproj-model-f16.gguf --image your_image.jpg -p "Describe this image."

About AutoRound

AutoRound is an advanced quantization technique from Intel that aims to minimize accuracy loss through automated rounding optimization.