Local Llm Vram Calculator

Local Llm Vram Calculator

Estimates the GPU memory a language model needs: weights, KV cache and total VRAM. For people who run models on their own graphics card and want to check the fit first.

Made by abdulrahmanmaslmany 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: Local Llm Vram Calculator estimates how much GPU memory a language model needs to run on your own machine. It works out the memory for the model weights, the KV cache and the total VRAM. The whole app is one Python file, app.py.

Who it suits: People who run language models on their own GPU and want to check a model fits before downloading it, and developers sizing hardware for a model.

Industry | Who uses it | How they use it | Example
Local AI | Hobbyists | Check whether a model fits their graphics card | Looking up the memory a model needs before downloading it
Software | Developers | Plan GPU memory for a model they will serve | Comparing the weight and KV cache shares of the total
IT | Infrastructure teams | Size hardware for a language model | Estimating the card a planned model needs

You can skip it if you only use hosted models, since you never manage the memory yourself.

Where to use it:

• A web browser, on the Space's own page at Hugging Face
• A copy on your own computer, using Python 3.12 on Windows, macOS or Linux

Use case | Who | What they get back
Check a model's fit | Local AI users | An estimate of weights, KV cache and total VRAM
Plan hardware | Developers and IT teams | A memory figure to compare against a GPU

How to set it up:

• A browser, if you only use the Space online
• Python 3.12 and pip, if you run a copy yourself

1. Open the Space's page on Hugging Face to use it online. Nothing needs installing.
2. For a local copy, install the huggingface_hub package and download the Space's files.
3. Move into the new folder and install Gradio. The repository has no requirements file.
4. Start the app with python.

pip install -U huggingface_hub
hf download abdulrahmanmaslmany/local-llm-vram-calculator --repo-type space --local-dir local-llm-vram-calculator
cd local-llm-vram-calculator
pip install gradio==6.30.0

How to use it: Open the Space's page, enter the details of the model you plan to run, and read the memory estimate it gives. The source does not list the input fields, so the page shows them. On your own computer, start the app first and open the page it serves.

python app.py

Tool | What it does
python app.py | Starts the calculator on your own computer

Example: You enter the details of a model you want to run locally. The page returns the memory its weights need, the memory its KV cache takes and the VRAM you should expect to use.

For newcomers and experts: If you are new to this, VRAM is the memory on your graphics card. Open the Space, try one model you know and compare the result with your card.

For experts, the app separates the weights from the KV cache, which grows with context length and is often what pushes a model past a card's limit. The repository is just README.md and app.py, pinned to Gradio 6.30.0.

Open the Space or run app.py
Enter the model's details
The app computes weights and KV cache
It adds them into a VRAM total
The estimate shows on the page

Good to know:

• Built with Gradio 6.30.0 and Python 3.12, and the main file is app.py
• License: mit
• Cost to run: none named
• Data it reaches: none named
• Real use can differ from any estimate, so leave some memory spare

On Hugging Face, 13 November 2026

• Likes: 1
• Built with: gradio
• License: mit

Links

• Space: https://huggingface.co/spaces/abdulrahmanmaslmany/local-llm-vram-calculator
• Files: https://huggingface.co/spaces/abdulrahmanmaslmany/local-llm-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.