https://github.com/neka-nat/cad3dify

2D to 3D CAD Conversion Using VLM
https://github.com/neka-nat/cad3dify

3d 3dcad ai cad drawing gpt-4o vlm

Last synced: 3 months ago
JSON representation

2D to 3D CAD Conversion Using VLM

• Host: GitHub
• URL: https://github.com/neka-nat/cad3dify
• Owner: neka-nat
• License: mit
• Created: 2022-09-18T10:42:06.000Z (almost 3 years ago)
• Default Branch: main
• Last Pushed: 2025-02-26T10:27:48.000Z (4 months ago)
• Last Synced: 2025-03-26T23:05:21.190Z (3 months ago)
• Topics: 3d, 3dcad, ai, cad, drawing, gpt-4o, vlm
• Language: Python
• Homepage:
• Size: 163 KB
• Stars: 64
• Watchers: 1
• Forks: 12
• Open Issues: 1
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

        
# cad3dify

Using GPT-4o (or Claude 3.5 sonnet, Gemini 2.0 flash, Llama 3.2 on Vertex AI), generate a 3D CAD model (STEP file) from a 2D CAD image.

## Getting started

Installation.

```bash

git clone [email protected]:neka-nat/cad3dify.git

cd cad3dify

poetry install

```

Run script.

A STEP`file ("output.step") will be generated.

```bash

cd scripts

export OPENAI_API_KEY=

python cli.py <2D CAD Image File>

```

Or run streamlit spp

```bash

streamlit run scripts/app.py

streamlit run scripts/app.py -- --model_type claude  # Use Claude 3.7 sonnet

streamlit run scripts/app.py -- --model_type gemini  # Use Gemini 2.0 flash

streamlit run scripts/app.py -- --model_type llama  # Use Llama 3.2 on Vertex AI

```

## Demo

We will use the sample file [here](http://cad.wp.xdomain.jp/).

### Input image

![input](sample_data/g1-3.jpg)

### Generated 3D CAD model

![output](sample_data/gen_result1.png)