Baguettotron WebGPU

Baguettotron WebGPU

Runs the small Baguettotron reasoning model in the browser through WebGPU, with no server and nothing to install. Suits people who want to try an in-browser model or read how it is wired.

Made by webml-community 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 Space runs Baguettotron, a small reasoning model from PleIAs, inside your browser. The name points to WebGPU, and the Space names an ONNX version of the model from the onnx-community group. It is a static web page, so no server runs the model for you.

Who it suits: It suits people who want to try a small language model without an account, and web developers who want to read a Vite and TypeScript project that runs a model in the browser.

Industry | Who uses it | How they use it | Example
Software | Web developers | Study how a model is loaded and run in a page | Open the src folder and follow the loading code
Machine learning | Hobbyists | Try a small reasoning model without any setup | Open the page and ask a short question
Education | Students | See an AI model run on their own device | Compare a few prompts and note the answers
Privacy-minded users | Anyone | Try a model that, by design, runs on their own machine | Read the page's notes on where it runs

Skip it if your browser lacks WebGPU or your device is low-powered, since the model runs on your own hardware.

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 dist/index.html in a web browser

Use case | Who | What they get back
Trying a small model | Anyone curious | Answers generated on your own device
Reading the code | Web developers | A Vite and TypeScript project with its src folder
Checking the models | Researchers | The two model repositories the page names

How to set it up: The online copy needs only a browser. To keep the files on your own computer, you need the following.

• A current browser, ideally one with WebGPU support
• Python with pip, used here to download the files
• The Hugging Face hub client, which the first step installs

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

pip install -U huggingface_hub
hf download webml-community/Baguettotron-WebGPU --repo-type space --local-dir Baguettotron-WebGPU
cd Baguettotron-WebGPU

How to use it: Open the Space's page, or load dist/index.html from your copy, and type a question into the page. The source does not describe the controls, so follow the labels on the page.

Example: You open the page and ask a short logic puzzle. The model works on it inside your browser and shows its answer on the page. The source gives no sample questions, timings or results.

For newcomers and experts: If you are new, start with a short question and give the page time to prepare the model on its first load. Later questions should be quicker.

If you know the field, the page pairs an ONNX build of the model with a browser runtime that uses WebGPU, so inference runs on your own graphics hardware. The project is built with Vite and TypeScript. Speed and memory use depend on your device, and the source does not give figures.

You open the page
The browser loads the ONNX model
WebGPU runs the model on your device
The answer appears on the page

Good to know: The source is short, so model size, speed and limits are not described.

• License: apache-2, as listed on Hugging Face
• Cost to run: no API key or paid service is named, and it uses your own device
• Data it reaches: the page says nothing about it, and the model runs in your browser, so check the page for what it downloads
• Not for: the source lists no exclusions, so check any answer before you rely on it

On Hugging Face, 13 November 2026

• Likes: 12
• Built with: static
• License: apache-2.0

Links

• Space: https://huggingface.co/spaces/webml-community/Baguettotron-WebGPU
• Files: https://huggingface.co/spaces/webml-community/Baguettotron-WebGPU/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.