Mindket

3D World Generator WebGPU

Newversion 1 · updated

Overview

Turn a text description into an interactive 3D world, delivered as one HTML file that uses Three.js. A small coding model runs in your browser through WebGPU.

3D World Generator WebGPU is a kit in the Developer category, listed by Mindket. It is free. This is version 1, last updated on . Download it on this page, a zip of its files. Nothing in it runs by itself.

At a glance

You need
  • Python on Windows, macOS or Linux
  • The pip command
  • A browser with WebGPU for the in-browser model
Runs in
  • A web browser with WebGPU, on the Space's page at Hugging Face
  • A local copy on Windows, macOS or Linux, running Python
Start with
pip install -U huggingface_hub
pip install gradio
python app.py

Platform

WindowsOperating system
macOSOperating system
LinuxOperating system

About

What it is

This Space turns a text description into an interactive 3D world. The result is a self-contained HTML page built with Three.js, the JavaScript library for 3D graphics in the browser.

A 4B-size coding model named Qwen3.5-4B-3D-Coder writes that page. The Q4_K_M build of it runs in your browser through WebGPU and the wllama library, so the generating happens on your own device.

Who it suits

It suits people who want to turn an idea into a small 3D scene without writing the code, and developers who want to see an in-browser language model produce Three.js pages.

IndustryWho uses itHow they use itExample
Game developmentIndie developersDescribe a scene and look at the generated worldSketching a level idea before building it
Web developmentFront-end developersUse the generated HTML as a starting pointStudying how a Three.js page is laid out
EducationTeachers and studentsShow how text becomes 3D codeDemonstrating what a small coding model can do
AI developmentDevelopersRead app.py to see how the Space loads the modelTrying the same approach with another model

Skip it if you need detailed, production-ready 3D scenes, since a small model writes one page at a time.

Where to use it

The app runs in these places.

  • A web browser with WebGPU, on the Space's page at Hugging Face
  • A local copy on Windows, macOS or Linux, running Python
Use caseWhoWhat they get back
Generate a 3D world from textAnyone curiousAn interactive Three.js page
Try the in-browser modelDevelopersA language model running on WebGPU
Study the SpaceLearnersA two-file repository with app.py and a README

How to set it up

Online you only need a browser that supports WebGPU. A local copy needs these.

  • Python on Windows, macOS or Linux
  • The pip command
  • A browser with WebGPU for the in-browser model
  1. Install the huggingface_hub package.
  2. Download the Space's files into a folder with the hf command.
  3. Move into that folder.
  4. Install gradio.
  5. Start the app with python app.py.
pip install -U huggingface_hub
pip install gradio
python app.py

Put your own value in its key setting first.

How to use it

Open the Space's page, or run the command below in your copy and open the page the app serves. Then describe the 3D world you want and run it.

python app.py

Example

You open the Space and describe a small scene, such as a field with a few trees. The app generates an interactive Three.js page for it, which you can look around in the browser.

For newcomers and experts

If you are new, open the Space's page in a browser that supports WebGPU and describe a simple scene first. A short description keeps the model's job small.

If you know the field, the model is Qwen3.5-4B-3D-Coder in a split GGUF build, quantised to Q4_K_M and loaded by wllama. Inference runs on the client through WebGPU, so the server only hosts the page. The trade-off is a large model download on first use and a limit set by your device.

  1. Open the Space page in a WebGPU browser
  2. Describe the 3D world in text
  3. The model writes a Three.js HTML page
  4. The world shows and you can explore it

Good to know

The repository has three files in all, with README.md and app.py at the top, and the Space is listed on zero-a10g hardware. The README says the model itself runs in your browser.

  • License: apache-2
  • Cost to run: none named
  • Data it reaches: the model repositories it names. The source says nothing else.

README and files

README

3D World Generator WebGPU

Turn a text description into an interactive 3D world, delivered as one HTML file that uses Three.js. A small coding model runs in your browser through WebGPU.

Made by ReallyFloppyPenguin 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 turns a text description into an interactive 3D world. The result is a self-contained HTML page built with Three.js, the JavaScript library for 3D graphics in the browser.

A 4B-size coding model named Qwen3.5-4B-3D-Coder writes that page. The Q4_K_M build of it runs in your browser through WebGPU and the wllama library, so the generating happens on your own device.

Who it suits: It suits people who want to turn an idea into a small 3D scene without writing the code, and developers who want to see an in-browser language model produce Three.js pages.

Industry | Who uses it | How they use it | Example
Game development | Indie developers | Describe a scene and look at the generated world | Sketching a level idea before building it
Web development | Front-end developers | Use the generated HTML as a starting point | Studying how a Three.js page is laid out
Education | Teachers and students | Show how text becomes 3D code | Demonstrating what a small coding model can do
AI development | Developers | Read app.py to see how the Space loads the model | Trying the same approach with another model

Skip it if you need detailed, production-ready 3D scenes, since a small model writes one page at a time.

Where to use it: The app runs in these places.

  • A web browser with WebGPU, on the Space's page at Hugging Face
  • A local copy on Windows, macOS or Linux, running Python
Use case | Who | What they get back
Generate a 3D world from text | Anyone curious | An interactive Three.js page
Try the in-browser model | Developers | A language model running on WebGPU
Study the Space | Learners | A two-file repository with app.py and a README

How to set it up: Online you only need a browser that supports WebGPU. A local copy needs these.

  • Python on Windows, macOS or Linux
  • The pip command
  • A browser with WebGPU for the in-browser model
  1. Install the huggingface_hub package.
  2. Download the Space's files into a folder with the hf command.
  3. Move into that folder.
  4. Install gradio.
  5. Start the app with python app.py.
pip install -U huggingface_hub
pip install gradio
python app.py

Put your own value in its key setting first.

How to use it: Open the Space's page, or run the command below in your copy and open the page the app serves. Then describe the 3D world you want and run it.

python app.py

Example: You open the Space and describe a small scene, such as a field with a few trees. The app generates an interactive Three.js page for it, which you can look around in the browser.

For newcomers and experts: If you are new, open the Space's page in a browser that supports WebGPU and describe a simple scene first. A short description keeps the model's job small.

If you know the field, the model is Qwen3.5-4B-3D-Coder in a split GGUF build, quantised to Q4_K_M and loaded by wllama. Inference runs on the client through WebGPU, so the server only hosts the page. The trade-off is a large model download on first use and a limit set by your device.

Open the Space page in a WebGPU browser
Describe the 3D world in text
The model writes a Three.js HTML page
The world shows and you can explore it

Good to know: The repository has three files in all, with README.md and app.py at the top, and the Space is listed on zero-a10g hardware. The README says the model itself runs in your browser.

  • License: apache-2
  • Cost to run: none named
  • Data it reaches: the model repositories it names. The source says nothing else.

On Hugging Face, 13 November 2026

  • Likes: 6
  • Built with: gradio
  • License: apache-2.0

Links

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.

Files
README.md  4 KB

1 file, 4 KB, downloaded as a zip.

What’s new

  • Version 1 · 2026-11-12

Compare with similar

Listed Hugging Face Spaces in the same category that share the most tags with this one.

ProductLikesBuilt withLicenseLast change
3D World Generator WebGPU This page6gradioapache-2.0
FLUX.1 [dev]Shares gradio9,583gradiomit
InstantIDShares gradio3,659gradioapache-2.0
Chat With Janus-Pro-7BShares gradio2,036gradiomit
PhotoMakerShares gradio1,960gradioapache-2.0
Z Image TurboShares gradio1,940gradioapache-2.0
Numbers as Mindket last read them, on 2026-11-12.

Reviews

No ratings yet

Write a review

No reviews yet.

More from Mindket

See all from Mindket

anaksor-ocr

Sokheng · Kit

Reads Khmer and English text from documents, receipts and photos of signs, using a detector for text lines and a recogniser for each line. Suits developers who…

New

Qwen2.5-7B-Instruct as GGUF files in eight quantization sizes, to chat with locally through llama.cpp. For people who want a capable 7.6B model on their own co…

New

ACE-Step 1.5 XL SFT

ACE-Step · Kit

Generates music from text with a 4B diffusion decoder, 50 steps and guidance control, aiming for higher audio quality than the base. For creators with a GPU of…

New

Similar in Developer

See all in Developer

Claude Code Skills & Plugins

alirezarezvani · Skill

Adds 411 ready-made skills for Claude Code and other coding agents, grouped by role from engineering and product to marketing, finance and compliance. Installs…

New29.8k

buzz

block · Kit

Self-hostable workspace where people and AI agents share the same rooms, built on a Nostr relay that logs every action as a signed event. For teams that want t…

New37.3k

plannotator

backnotprop · Kit

Opens your AI coding agent's plans, specs and code changes in a browser so you can annotate or comment on them and send the feedback straight back to the agent.

New10.5k