GGUF Model VRAM Calculator

GGUF Model VRAM Calculator

Estimates how much video memory a GGUF model needs, in a static page that runs in your browser. Suits people choosing a quantized model that fits their graphics card.

Made by DavidAU on Hugging Face. These notes are Mindket's, written from its Space page and README; the app's own files and license are on Hugging Face.

About

What it is: This is a static web page on Hugging Face Spaces for working out how much video memory a GGUF model needs. It is a copy of an earlier calculator by NyxKrage, adapted by DavidAU. You open it in a browser, so there is nothing to install. The source does not list its input fields.

Who it suits: It suits people who run language models on their own graphics card and want to check the fit before they download a large file.

Industry | Who uses it | How they use it | Example
Local AI | Hobbyists | Check whether a quantized model fits their card before downloading it | Look up the memory a 7B model needs
IT and hardware | Engineers sizing machines | Compare memory needs across models when planning a GPU purchase | Compare two model sizes side by side
Education | Students learning quantization | See how model size relates to memory use | Try several models and note the differences

Skip it if you need a measured number from a real run, since a calculator gives estimates.

Where to use it:

• A browser: open the Space's page on Hugging Face, with nothing to install
• A downloaded copy on Windows, macOS or Linux, opened from index.html

Use case | Who | What they get back
Sizing a model | Local model users | An estimate of the video memory a GGUF file needs
Planning hardware | Engineers | A way to compare memory needs before buying
Reading the page | Web developers | A small static project with an HTML file and style sheets

How to set it up: The online copy needs nothing. To keep the files on your own computer, you need the following.

• Python with pip, used here to download the files
• The Hugging Face hub client, which the first step installs
• Any current web browser to open the page

1. Install the Hugging Face hub client.
2. Download the Space's files into a local folder.
3. Move into that folder.
4. Load index.html in your browser.

pip install -U huggingface_hub
hf download DavidAU/GGUF-Model-VRAM-Calculator --repo-type space --local-dir GGUF-Model-VRAM-Calculator
cd GGUF-Model-VRAM-Calculator

How to use it: Open the Space's page, or load index.html from your copy, then fill in the details of the model you are considering. The source does not describe the fields, so follow the labels on the page and read the estimate it shows.

Example: You are choosing between two quantized versions of a 7B model for a card with limited memory. You enter each one's details in turn and compare the two estimates the page gives. The source does not say which fields it asks for.

For newcomers and experts: If you are new, a GGUF file is a packaged language model that you run on your own machine, and it needs video memory to load. Open the page and try the model the Space names, Starling-LM-7B-beta, to see what an estimate looks like.

If you know the field, the repository holds one HTML file and two style sheets, so the arithmetic most likely runs in your browser with no server. The source does not say which factors it counts, so check any result against your own load.

You open the page
You enter the model's details
The page works out the memory
You read the estimate

Good to know: The source is short, so what the calculator counts is not documented here.

• License: mit, as listed on Hugging Face
• Cost to run: no API key or paid service is named
• Data it reaches: none named, since it is a static page
• Not for: the source lists no exclusions, so treat results as estimates and check them against your own runs

On Hugging Face, 13 November 2026

• Likes: 100
• Built with: static
• License: mit

Links

• Space: https://huggingface.co/spaces/DavidAU/GGUF-Model-VRAM-Calculator
• Files: https://huggingface.co/spaces/DavidAU/GGUF-Model-VRAM-Calculator/tree/main

Try it

Open the Space in your browser, or copy it to your own Hugging Face account from its page. Nothing in this kit runs by itself.