gemma-3-270m-it-GGUF

gemma-3-270m-it-GGUF

Generates text from prompts as a 268M instruction-tuned Gemma 3 model in GGUF format, small enough for laptops and desktops. For developers who want a tiny local model to run or fine-tune.

Made by unsloth 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-3-270m-it-GGUF is Unsloth's GGUF copy of Google's instruction-tuned Gemma 3 model with 268 million parameters. It reads a text prompt and writes text back, such as an answer or a summary. The card offers a free fine-tuning notebook for this size.

Who it suits: It suits developers and learners who want a tiny model on their own computer, and people who want a cheap model to practice fine-tuning on.

Industry | Who uses it | How they use it | Example
Software | Developers | Run a small text model locally | Answering short questions inside a tool
Education | Learners | Practice fine-tuning on a tiny model | Following the free notebook for this size
Research | Experimenters | Test prompts and sampling settings on a small model | Comparing replies at different settings

You can skip it if you need the larger context of the bigger Gemma 3 sizes, which reach 128K tokens.

Where to use it:

• Runtimes that load GGUF files
• Google Colab, through the free notebook the card links for fine-tuning
• Windows, macOS or Linux, with Python for the download step

Use case | Who | What they get back
Question answering | Developers | A short written answer to your prompt
Summarizing | Writers and developers | A shorter version of a document you paste in
Fine-tuning practice | Learners | A small model to adapt to your own examples

How to set it up:

• Python, to run the Hugging Face download tool
• The huggingface_hub package
• A runtime that can load GGUF files
• Room for about 134 MB of model file, as a 4-bit estimate

1. Install the Hugging Face hub package.
2. Download the repository files to your computer.
3. Load the gguf file in your runtime and apply the sampling settings below.

pip install -U huggingface_hub
hf download unsloth/gemma-3-270m-it-GGUF

The card asks you to use these sampling settings.

temperature = 1.0, top_k = 64, top_p = 0.95, min_p = 0.0

How to use it: You load the model in your GGUF runtime, set the sampling values above, and type a prompt. The card also points to an Unsloth guide on running Gemma 3 correctly.

Tool | What it does
hf download | Fetches the model files to your computer
Sampling settings | The values in the code block above, which the card says to use
Unsloth notebook | A free fine-tuning notebook for this model size

Example: You paste a paragraph and ask, "Summarize this in two sentences." The model writes a short summary back as plain text.

For newcomers and experts: If you are new, this model is tiny and fits modest hardware. Download the files, load them in a runtime, apply the settings above, and try a short question.

If you know the field, it is the instruction-tuned 270M member of the Gemma 3 family with a 32K-token input context, converted to GGUF. The trade-off is size, since the family runs up to 27B.

You type a prompt
The runtime loads the GGUF file
The model samples with the card's settings
Text comes back

Good to know:

Title: Input context by Gemma 3 size, in thousand tokens
270M and 1B sizes: 32 K tokens
4B, 12B and 27B sizes: 128 K tokens

• Size: about 134 MB at 4-bit as an estimate
• Family: Gemma 3 from Google DeepMind, with multilingual support in over 140 languages
• License: gemma, under Google's Gemma terms
• Cost to run: none named
• Data it reaches: the prompts you give it, and the files you download from Hugging Face
• Not for: the shown card text names none, but the Gemma terms of use apply

On Hugging Face, 13 November 2026

• Task: text generation
• Library: transformers
• Parameters: 268M
• Weights: about 134 MB at 4-bit (estimate)
• Formats: GGUF
• Downloads in the last 30 days: 91,489
• Likes: 181
• License: gemma
• Based on: google/gemma-3-270m-it
• Quantized versions: 2
• Tags: transformers, gguf, gemma3, unsloth, gemma, google, text-generation, conversational

Links

• Model: https://huggingface.co/unsloth/gemma-3-270m-it-GGUF
• Files: https://huggingface.co/unsloth/gemma-3-270m-it-GGUF/tree/main

Get the model

Download its files with Hugging Face's command line:

pip install -U huggingface_hub
hf download unsloth/gemma-3-270m-it-GGUF

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