WAM DiT4DiT — RoboCasa Kitchen / Wan2.2 — 3latin2latout
Video-only (world-model) finetune of Wan2.2-TI2V-5B on RoboCasa Kitchen.
Configuration
| training_mode | video (no action head — action DiT is not trained) |
| latent conditioning | 3latin2latout (cond latent frames -> 2 future latent frames) |
| video EMA | saved (_video_ema_model.*, 825 tensors, fp32) |
| EMA schedule | diffusers use_ema_warmup (inv_gamma=1.0, power=0.75, cap 0.9999) |
| extraction | WAM_VIDEO_EMA_TRAIN_EXTRACT=0 — EMA is tracked+saved only, training math unchanged |
| GPUs | 8 x H200 |
Contents
Milestone checkpoints are stored per step under checkpoint-<step>/.
Optimizer state (global_step*) is intentionally excluded, so these checkpoints are
for inference / feature extraction / probing only — they cannot resume training.
Each folder contains the model shards (model-*.safetensors + index) and processor/config
files. The EMA copy of the video DiT is included in the same shards under the
_video_ema_model.* prefix.
Base model & license
Derived from Wan-AI/Wan2.2-TI2V-5B-Diffusers; the base model's license terms apply to this derivative.
- Downloads last month
- 24
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support
Model tree for hmkang/wam-dit4dit-robocasa-wan22-3latin2latout
Base model
Wan-AI/Wan2.2-TI2V-5B-Diffusers