princeton-vl

princeton-vl / procfunc

Public

Function-Oriented Abstractions for Procedural 3D Generation in Python

17
0
100% credibility
Found May 01, 2026 at 17 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
Python
AI Summary

ProcFunc offers Python functions and a transpiler to simplify procedural 3D modeling and shading in Blender.

How It Works

1
🔍 Discover easy 3D magic

You hear about a simple way to create endless 3D scenes in Blender using friendly Python code.

2
📦 Quick setup

Install the tool with one easy command and open Blender.

3
📖 Follow welcoming guides

Check out ready examples and clear instructions to get inspired.

4
✨ Turn Blender art into code

Pick a Blender file, and instantly get editable Python code that recreates it.

5
✏️ Tweak your creation

Change numbers or add ideas in the simple code to customize your scene.

6
▶️ Run and watch it build

Hit run, and see your procedural 3D world come alive in Blender.

🎉 Endless 3D worlds unlocked

Now you generate infinite variations of stunning 3D art effortlessly!

Sign up to see the full architecture

5 more

Sign Up Free

Star Growth

See how this repo grew from 17 to 17 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is procfunc?

Procfunc brings function-oriented abstractions to procedural 3D generation in Python, letting you build geometry nodes, shaders, and meshes as composable functions instead of visual graphs. It transpiles Blender .blend files into executable Python code and runs everything inside Blender via its Python API (bpy 4.2). Developers get clean, readable scripts for generating complex 3D assets programmatically.

Why is it gaining traction?

Unlike raw Blender scripting or node editors, procfunc offers a functional API that feels like writing shaders in GLSL—chain math ops, noises, and BSDFs into reusable functions. The transpiler bridges visual prototyping to code, and built-in tracing/debugging makes iteration fast. Early LLM experiments show it excels at AI-driven asset generation, where code output directly renders in Blender.

Who should use this?

Blender Python scripters building procedural terrains, plants, or materials for games/tools. AI researchers fine-tuning LLMs on 3D prompts for geometry editing or from-scratch generation. Teams needing version-controlled procedural pipelines over static node trees.

Verdict

Grab it if you're prototyping LLM 3D gen—docs are solid, install is pip-simple, but pin to <0.31 due to pre-1.0 flux and 17 stars signal early days (1.0% credibility). Experiments folder has repro steps; main lib delivers on promises today.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.