gemma-4-E4B-it

gemma-4-E4B-it

Handles text, image and audio input and writes text, with a 128K-token context and support for over 140 languages. A small instruction-tuned Gemma 4 model from Google DeepMind, built to run on laptops and phones.

Made by google 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 is the instruction-tuned E4B model in Google DeepMind's Gemma 4 family of open models. It takes text, images and audio as input and writes text as output. It has 4.5B effective parameters, or 8B counting embeddings, and the card says it is designed for efficient local use on laptops and mobile devices.

Who it suits: Developers who want a small open model for chat, coding, reasoning and multimodal tasks that can run on modest hardware.

Industry | Who uses it | How they use it | Example
Software development | Developers | Add a local model for text, code and reasoning | A coding helper that runs on a laptop
Mobile and edge apps | App builders | Run a model on the device itself | An assistant that works on a phone
Accessibility and media | Product teams | Work with images and spoken audio | A model that reads an image and answers about it
Global products | Multilingual teams | Serve users in many languages | Replies in one of the 140-plus supported languages

If you need the largest quality at any cost, you can skip it, because the family also has 12B, 26B A4B and 31B models.

Where to use it:

• Python with the transformers library on Windows, macOS or Linux
• Laptops and mobile devices, which the card names as targets for the small models
• Servers and workstations, where larger Gemma 4 sizes are the card's target

Use case | Who | What they get back
Chat and coding | Developers | Text answers and code from a small model
Image questions | App builders | Text answers about an image you supply
Audio input | Product teams | Text output from audio, which E4B supports

How to set it up:

• Python and pip
• About 16 GB of disk space in BF16, or about 4 GB at 4-bit (an estimate)
• The transformers library to load the files

1. Install the Hugging Face hub tools.
2. Download the model files.
3. Load the model with transformers, following the model card.

pip install -U huggingface_hub
hf download google/gemma-4-E4B-it

How to use it: Send the model text, an image or audio and read the text it writes back. The model supports a system role for more controlled conversations, and thinking modes you can configure. It also has built-in support for function calling, which agents use to call tools.

Example: You send a photo along with a short question about what it shows. The model reads the image and writes a text answer. You can add a system message to set its tone or role first.

For newcomers and experts: If you are new to this, start with a short text question and a system message, then add an image. The E4B size is the one aimed at phones and laptops.

The E in E4B means effective parameters. The small models use per-layer embeddings, which are big tables consulted only for quick lookups. The model has 42 layers, a 512-token sliding window mixed with global attention, a 262K vocabulary, and encoders of about 150M for vision and about 300M for audio.

You send text, an image or audio
The model reads it, with thinking if enabled
It writes text as output
You use the answer or call a tool

Good to know:

• Context: the small models handle 128K tokens, and the medium ones handle 256K
• Output: it generates text only, even when the input is an image or audio

Title: Total parameters of the Gemma 4 dense sizes, in billions
E2B: 2.3 billion
E4B: 4.5 billion
12B: 11.95 billion
31B: 30.7 billion

• License: apache-2.0
• Cost to run: no API key or paid service is named, only your own hardware
• Data it reaches: downloading fetches model files, and it reads the text, images and audio you give it
• Not for: producing images or audio, since the output is text

On Hugging Face, 13 November 2026

• Task: any to any
• Library: transformers
• Parameters: 8B
• Weights: about 16 GB in BF16; about 4 GB at 4-bit (estimate)
• Formats: Safetensors
• Downloads in the last 30 days: 4,298,348
• Likes: 1,850
• License: apache-2.0
• Based on: google/gemma-4-E4B
• Quantized versions: 450
• Fine-tunes: 473
• Tags: transformers, safetensors, gemma4, image-text-to-text, any-to-any

Links

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

Get the model

Download its files with Hugging Face's command line:

pip install -U huggingface_hub
hf download google/gemma-4-E4B-it

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