https://github.com/soniccodes/lucid-v1
realtime latent world model inference demo
https://github.com/soniccodes/lucid-v1
diffusion-models generative-model machine-learning minecraft
Last synced: 3 days ago
JSON representation
realtime latent world model inference demo
- Host: GitHub
- URL: https://github.com/soniccodes/lucid-v1
- Owner: SonicCodes
- Created: 2024-11-11T14:02:43.000Z (over 1 year ago)
- Default Branch: master
- Last Pushed: 2024-11-11T14:45:25.000Z (over 1 year ago)
- Last Synced: 2025-11-18T12:08:31.549Z (7 months ago)
- Language: Python
- Homepage: https://lucidv1-demo.vercel.app/
- Size: 11.4 MB
- Stars: 48
- Watchers: 2
- Forks: 5
- Open Issues: 2
-
Metadata Files:
- Readme: ReadMe.md
Awesome Lists containing this project
README
# Lucid v1, "latent world model"
This is an inference script with full interactive realtime play (and web ui), it contains a couple of maps and more will be added soon, this is compatible with RTX 4090-grade gpus (upto 30fps), you can adjust context length and sampling steps to increase fidelity, optimized/default setting now should run at 20fps
## How to install
### Ubuntu / Debian
#### Prerequisite
```bash
sudo apt-get install libgl1 unzip
```
Please make sure to have opencv, unzip, and nvidia cuda driver installed, [use this link to install cuda](https://developer.nvidia.com/cuda-12-0-0-download-archive)
#### Installation
```bash
sudo chmod +x ./setup.sh
./setup.sh
```
the setup script should install all the necessary files, pls check nvidia-smi again before proceeding, and currently i'm releasing support only for cuda 12.0
#### Launch
```bash
python3 launch.py --port 3000
```