Qwen3.5-4B-GLB-3D-Coder
What it is
A 4-billion parameter text generation model fine-tuned specifically for generating 3D content, including GLB files, Three.js code, and Godot scripts. It is distributed as a GGUF file on Hugging Face and is designed to be run locally using inference engines like llama.cpp, vLLM, or Ollama.
The problem it solves
The lack of specialized, locally-runnable LLMs that can directly output valid 3D code and file formats (like GLB) for game and web development.
What you could use it for
- Generating 3D model files in GLB format from text descriptions
- Writing Three.js code for web-based 3D graphics
- Creating Godot engine scripts for 3D game development
Why it matters
Enables developers and creators to generate 3D assets and code locally without relying on cloud-based APIs, lowering the barrier to entry for 3D content creation.
Read the source ↗Back to the library →
Summary written by an AI model from the source; check the source before you rely on it.