gemma-4-E4B-it-GGUF by ggml-org

gemma-4-E4B-it-GGUF by ggml-org

A GGUF-format copy of Google's gemma-4-E4B-it that you can run on your own computer with the Llama app or the llama serve command. For people who want Gemma running locally.

Made by ggml-org on Hugging Face. These notes are Mindket's, written from its model page and model card; the model's own files and license are on Hugging Face.

About

What it is: gemma-4-E4B-it-GGUF is a GGUF-format copy of Google's gemma-4-E4B-it, published by ggml-org. It was converted automatically from the original models, so it is the same model packaged for Llama tools. The record lists about 7.5B parameters and tags it as any to any, and the card says little beyond how to run it.

Who it suits: People who want to run Gemma on their own machine through the Llama app or the llama serve command, and developers who test GGUF models.

Industry | Who uses it | How they use it | Example
Software | Application developers | Serve a local model for an app to call | Start the model with the llama serve command
Hobby computing | Local AI enthusiasts | Try Gemma on their own computer without a cloud service | Open the model in the Llama app
Education | Students | Learn how GGUF models are fetched and loaded | Pull the files with hf download and look at them

Anyone who needs benchmarks, quantization details or usage notes can skip it, since the card is nearly empty.

Where to use it: It runs locally, in the Llama app or from a terminal.

• The Llama app
• A terminal, with the llama serve command
• Windows, macOS or Linux with Python, for downloading the files with hf

Use case | Who | What they get back
Local chat | Individuals | Replies from a model running on their own computer
Local serving | Developers | A running llama serve instance loaded with this model

How to set it up: You can let llama serve fetch the model, or download the files first.

• The Llama app or the llama command
• Free disk space of a few GB, since the 4-bit weight estimate is 3.8 GB
• Python with pip, only if you download with hf

1. Get the Llama app or the llama command.
2. Run the serve command, which pulls the model from Hugging Face.
3. Or download the files yourself with the hf command.

llama serve -hf ggml-org/gemma-4-E4B-it-GGUF

pip install -U huggingface_hub
hf download ggml-org/gemma-4-E4B-it-GGUF

How to use it: Day to day you launch the model with llama serve or open it in the Llama app. The card documents no other commands or settings.

Tool | What it does
llama serve with the -hf flag | Loads the model from Hugging Face and serves it
hf download | Copies the model files to your computer

Example: You run the serve command in a terminal with this repository name. The model loads from Hugging Face and starts serving. The card shows no sample chat, so what you ask it next is up to you.

For newcomers and experts: GGUF is a file format for running a model on your own machine. Think of this as a ready-to-run package of Google's Gemma. The first small thing to try is opening the Llama app and loading this model, or running the serve command once.

It is an automatic conversion of google/gemma-4-E4B-it. The card lists four source models, including an assistant variant and two QAT q4_0 unquantized ones. It does not say which quantization levels the repository holds. Its to-do list mentions adding info and dflash, so expect the card to change.

Start from the original Gemma model
Convert it to GGUF automatically
Serve it with llama or download it
Chat with it on your own computer

Good to know: The card is a stub, so most details come from the repository record.

• Thin card: it gives no benchmarks, quantization list or usage notes.
• Source: converted automatically from google/gemma-4-E4B-it.
• Size: about 7.5B parameters, with a 4-bit weight estimate of 3.8 GB.
• License: Apache 2.0.
• Cost to run: none named.
• Data it reaches: the model files fetched from Hugging Face. The card names no account or other access.

On Hugging Face, 13 November 2026

• Task: any to any
• Parameters: 7.5B
• Weights: about 3.8 GB at 4-bit (estimate)
• Formats: GGUF
• Downloads in the last 30 days: 263,980
• Likes: 110
• License: apache-2.0
• Based on: google/gemma-4-E4B-it
• Quantized versions: 1
• Tags: gguf, quantized, any-to-any, conversational

Links

• Model: https://huggingface.co/ggml-org/gemma-4-E4B-it-GGUF
• Files: https://huggingface.co/ggml-org/gemma-4-E4B-it-GGUF/tree/main

Get the model

Download its files with Hugging Face's command line:

pip install -U huggingface_hub
hf download ggml-org/gemma-4-E4B-it-GGUF

Then follow the usage in its model card. Nothing in this kit runs by itself.