| --- |
| license: apache-2.0 |
| tags: |
| - world-model |
| - minecraft |
| - video-generation |
| - causal-forcing |
| --- |
| |
| # ForgeWM Checkpoints |
|
|
| Training checkpoints for [ForgeWM](https://github.com/asdfo123/ForgeWM). |
|
|
| ## Available Checkpoints |
|
|
| | File | Stage | Description | |
| |------|-------|-------------| |
| | `stage0/model.pt` | Stage 0 | Bidirectional SFT (domain adaptation, 4000 steps) | |
| | `stage1/model.pt` | Stage 1 | Teacher-Forcing Causal AR (10000 steps) | |
| | `stage2/model.pt` | Stage 2 | Consistency Distillation (6000 steps) | |
| | `stage3/model.pt` | Stage 3 | DMD final model (4-step real-time inference) | |
|
|
|
|
|
|
| ## Usage |
|
|
| ```bash |
| # Download |
| huggingface-cli download asdfo123/ForgeWM --local-dir ./ckpts --repo-type model |
| |
| # Inference |
| python inference.py --checkpoint_path ckpts/stage3/model.pt --image_path demo_images/forest.png --action_type forward |
| ``` |
|
|
| ## 🚧 Coming Soon |
| - More intermediate checkpoints |
|
|
| ## Citation |
|
|
| ```bibtex |
| @misc{forgewm2026, |
| title={ForgeWM: A Reproducible Training Recipe for Action-Controllable World Models}, |
| author=ForgeWM Team, |
| year={2026}, |
| url={https://github.com/asdfo123/ForgeWM} |
| } |
| ``` |
|
|
| ## Contact |
| [Xinye Li](https://asdfo123.github.io) leeasdfo123@gmail.com |
