NVIDIA has introduced Alpamayo 2 Super, a large-scale vision-language-action model designed for use in robotaxis and autonomous driving systems. This model is made available under the permissive OpenMDW-1.1 license, allowing for fine-tuning, derivative development, and commercial redistribution. Alpamayo 2 Super combines a 32-billion parameter Cosmos 3 Super Reasoner backbone with a 2.3-billion parameter diffusion action decoder. The model has demonstrated strong performance on certain benchmarks, including a score of 79.2 on LingoQA. Alpamayo 2 Super can generate a range of outputs, including trajectories, causal chains, and labeled visual data, from a single pass through the model. This development has implications for the advancement of autonomous vehicle technology.