| ## Explainability |
|
|
| | Field | Response | |
| | :---- | :---- | |
| | Intended Application & Domain | World generation and simulation for Physical AI, specifically closed-loop driving simulation and synthetic multi-sensor data generation. | |
| | Model Type | Transformer-based Mixture-of-Transformers (MoT) model, derived from Cosmos3-Nano through multi-sensor driving post-training and causal distillation. | |
| | Intended Users | Physical AI developers, researchers, and practitioners building or evaluating autonomous vehicle perception and driving-policy workflows. | |
| | Output | Synchronized multi-view camera videos and LiDAR observations. Camera outputs are MP4 videos or streamed frames. LiDAR outputs are range and intensity maps that can be converted to 3D point clouds with per-point intensity. | |
| | Tools used to evaluate datasets to identify synthetic data and ensure data authenticity. | Dataset provenance analysis, metadata validation, watermark and artifact detection, embedding-based clustering, heuristic quality checks, and model-assisted data validation pipelines are used to identify synthetic content patterns, assess dataset authenticity, and improve data quality during dataset curation. | |
| | Describe how the model works | The model generates synchronized multi-view camera videos and LiDAR observations conditioned on text and scene controls. It uses cached history and incoming controls to support autoregressive generation for closed-loop driving simulation. | |
| | List the groups or group characteristics for which this was tested to produce comparable outcomes | None. | |
| | Technical Limitations & Mitigation | The model may not follow input conditions accurately in challenging cases, especially those involving complex scene composition, unusual camera motion, multiple interacting agents, low lighting, high motion blur, or fine-grained physical interactions. Generated outputs may contain temporal inconsistency, object morphing, inaccurate 3D structure, or implausible physical dynamics. Users should validate output quality and adherence to input conditions for their intended use cases. | |
| | Verified to have met prescribed NVIDIA quality standards | Yes. | |
| | Performance Metrics | Video evaluation includes Frame FID, FVD (I3D and StyleGAN-V), DOVER technical, temporal Sampson error, temporal consistency (ACM), and cross-view Sampson errors (CSE2 and CSE5). LiDAR evaluation includes global and local BEV JSD, BEV MMD², Fréchet Utonia distance, Utonia MMD², temporal and vehicle CTC ratios, object support recall, and object detection F1. Driving-specific evaluation includes AutoGrader and human assessment. Please see the overview for the reported benchmark results. | |
| | Potential Known Risks | This model can generate synthetic media and may produce content that is offensive, unsafe, misleading, indecent, or unsuitable for a target deployment. Users should implement robust safety guardrails — including content filtering, abuse monitoring, and access controls — to reduce the risk of harmful outputs. Users are responsible for ensuring that their use of the model complies with all applicable laws and regulations, and for regularly reviewing and updating their guardrails as risks evolve. | |
| | Licensing | [OpenMDW1.1](https://openmdw.ai/license/1-1/) | |
|
|
