DreamDojo-AgiBot-14B-Diffusers

Large model fine-tuned on AgiBot. Part of the DreamDojo model family.

Size	14B
Stage	Post-training
Architecture	DiT (Diffusion Transformer) with AdaLN-LoRA
Base	Cosmos Predict 2.5

Checkpoint Structure

DreamDojo-AgiBot-14B-Diffusers/
├── transformer/            # DiT backbone (sharded safetensors)
├── crossattn_adapter/      # Text-to-DiT projection (100352 → 1024)
├── vae/                    # AutoencoderKLWan (standard diffusers)
├── lam/                    # Latent Action Model (710M params)
├── text_encoder/           # Cosmos-Reason1-7B
├── scheduler/              # FlowMatchEulerDiscreteScheduler
├── action_processor/       # DreamDojo-specific config
└── config.json

Architecture

	14B
Model channels	5120
Transformer blocks	36
Attention heads	40
Patch size (spatial / temporal)	2 / 1
Action dim	384 (unified)

Citation

@article{dreamdojo2025,
  title={DreamDojo: Advancing Real-World Robot Policies Through Generated Interactive Environments},
  author={NVIDIA},
  year={2025}
}

License

Please refer to the NVIDIA DreamDojo repository for license terms.