EasyWAM-MoT-Cosmos25

This repository hosts released EasyWAM-MoT checkpoints built on Cosmos-Predict2.5-2B. EasyWAM-MoT uses separate Video DiT and Action DiT experts whose tokens interact through mixed self-attention. At inference time, it predicts actions without generating future video. The checkpoints are trained with the EasyWAM codebase.

LIBERO

Results

Task success rate (%) under the EasyWAM LIBERO evaluation protocol. All results use state_position: sequence; higher is better. Models using the same Cosmos-Predict2.5-2B backbone are shown together for comparison, and 🔥 marks the architecture released in this repository.

Full-Parameter

Model	Spatial	Object	Goal	LIBERO-10	Avg.
EasyWAM-Unified	97.8	99.4	97.0	93.0	96.8
🔥 EasyWAM-MoT	98.0	98.4	98.4	95.6	97.6
EasyWAM-MoT-Joint	98.6	99.8	98.4	96.0	98.2
EasyWAM-MoT-IDM	99.4	99.4	99.8	98.4	99.3
EasyWAM-Hidden	97.2	98.8	96.0	95.0	96.8

LIBERO-Plus

Model	Background	Camera	Language	Layout	Light	Noise	Robot	Avg.
EasyWAM-Unified	72.7	63.7	90.1	82.6	89.2	72.1	79.2	78.2
🔥 EasyWAM-MoT	60.7	75.1	92.3	82.6	92.6	81.3	58.5	77.7
EasyWAM-MoT-Joint	62.1	60.8	96.6	84.5	94.4	69.4	76.9	77.7
EasyWAM-MoT-IDM	66.4	59.8	93.8	85.7	89.0	72.3	81.9	78.4
EasyWAM-Hidden	59.5	65.9	92.6	85.0	89.1	68.3	82.5	77.8

LIBERO-Plus results are reported for full-parameter checkpoints only. See the complete EasyWAM benchmark table for source results and comparisons across backbones.

Download

hf download OpenMOSS-Team/EasyWAM-MoT-Cosmos25 \
  libero_mot_cosmos25.pt \
  libero_dataset_stats.json \
  --local-dir ./checkpoints

Evaluation

Install EasyWAM, prepare the Cosmos-Predict2.5-2B dependencies using the backbone guide, and set up the simulator using the LIBERO evaluation guide. Then run:

# Full-parameter checkpoint
python experiments/libero/run_libero_manager.py \
  task=libero_easywam_mot_cosmos25 \
  ckpt=./checkpoints/libero_mot_cosmos25.pt \
  EVALUATION.dataset_stats_path=./checkpoints/libero_dataset_stats.json

The default protocol evaluates all four LIBERO suites with 50 trials per task. Add MULTIRUN.num_gpus=<gpu-count> to distribute evaluation across multiple GPUs.

Available Checkpoints

• LIBERO full-parameter checkpoint: libero_mot_cosmos25.pt
• LIBERO normalization statistics: libero_dataset_stats.json
• State-token placement: sequence
• Checkpoint format: EasyWAM PyTorch checkpoint (.pt)

Project

• Project page: http://openmoss.ai/EasyWAM/
• Code: https://github.com/OpenMOSS/EasyWAM

License and Citation

EasyWAM code is released under the MIT License. Released checkpoints in this repository use CC BY-NC 4.0 and remain subject to the terms of their base models and training data. If EasyWAM is useful in your research, please cite:

@misc{easywam2026,
  title  = {EasyWAM: A Unified and Efficient Framework for Training and Evaluating World Action Models},
  author = {EasyWAM-Team},
  year   = {2026},
  url    = {https://github.com/OpenMOSS/EasyWAM}
}