Downloads · 30 days
0
pixel-art-gwm
Aquinlabs/pixel-art-gwm
pixel-art-gwm is a machine learning model from Aquinlabs. Use it for the machine learning task on the model card, and read the license before you ship it in a product. It is set up for pytorch.
Pixel Art GWM is a compact action-conditioned world model for a 2D pixel-art platformer. It predicts how the game changes in response to actions such as moving, jumping, and using objects.
Access
Public
Updated Sep 21, 2026
Repo size
23.5 MB
Likes
0
Public
Repo makeup
Click a slice to open those files.
.pt52.9 MB · 100%
At a glance
- Library
- pytorch
- Access
- Public
- Created
- Sep 21, 2026
- Updated
- Sep 21, 2026
- SHA
- 35a6ee93
Model card
From the Hugging Face model README
Pixel Art GWM
Pixel Art GWM is a compact action-conditioned world model for a 2D pixel-art platformer. It predicts how the game changes in response to actions such as moving, jumping, and using objects.
Highlights
- 4.4M parameters
- Tokenizer and dynamics model included
- Supports imagined multi-step rollouts
- Runs locally with PyTorch
- MIT licensed
Files
models/tokenizer.pt: released tokenizermodels/dynamics.pt: released dynamics modelmodels/dyn_v3_26k/: latest checkpoint releasemodels/dyn_v2_10k/: earlier checkpoint release
Usage
python play_local.py --tok models/tokenizer.pt --dyn models/dynamics.pt
The original code is available at GitHub.
Dataset
License
MIT
- Library
- pytorch
- Created
- Sep 21, 2026
- Updated
- Sep 21, 2026