--- license: apache-2.0 tags: - game-agent - world-model - video-generation - pytorch --- # GameDirector checkpoints This repository contains the three trained checkpoint families used by GameDirector: AttackNet, SituationNet, and the causal autoregressive world model. The world-model checkpoint contains the trained DiT only; its Wan VAE, UMT5 encoder, and tokenizer must be downloaded separately. ## Checkpoint manifest | Path | Contents | |---|---| | `attacknet/attacknet.pt` | three-game hit detector | | `situationnet/situationnet.pt` | three-game spatial-state estimator | | `world_model/ar_world_model.safetensors` | causal AR world-model DiT used by the agent (9.4 GB) | `SHA256SUMS` lists the expected hashes; check them with `sha256sum -c SHA256SUMS`. ## Training data The public NRFTW and Hollow Knight training splits are available at [onepiece1999/GameDirector-Training-Data](https://huggingface.co/datasets/onepiece1999/GameDirector-Training-Data). The V Rising training split used by these three-game checkpoints is not included in that dataset. ## Download ```bash hf download onepiece1999/GameDirector --local-dir ./GameDirector-models ``` ## License The checkpoints are released under the Apache License 2.0.