NVIDIA Releases Alpamayo 2 Super: A 34B Open Vision-Language-Action Model for Robotaxis and Autonomous Driving Under OpenMDW-1.1
Evolving story · 2 updatesNVIDIA Alpamayo 2 Super for Autonomous VehiclesTimeline →NVIDIA launched Alpamayo 2 Super, a 34-billion-parameter open model for autonomous driving that combines vision, language, and action generation under a permissive license.
- Alpamayo 2 Super is a 34B open model combining vision, language, and action generation for autonomous driving.
- It uses a 32B reasoning backbone and a 2.3B diffusion action decoder to produce trajectories and explanations in one pass.
- The model is released under OpenMDW-1.1, allowing commercial use, fine-tuning, and derivative development.
- Scores 79.2 on LingoQA, demonstrating strong performance in autonomous driving vision-language tasks.
NVIDIA has introduced Alpamayo 2 Super, a 34-billion-parameter open model designed to power next-generation autonomous driving systems. The model integrates a 32B Cosmos 3 Super Reasoner backbone with a 2.3B diffusion-based action decoder, enabling it to process visual inputs, generate language explanations, and produce actionable driving trajectories in a single inference pass.
Under the OpenMDW-1.1 license, the model supports fine-tuning, derivative creation, and commercial redistribution, making it accessible for both research and enterprise applications. Key capabilities include Chain-of-Causation traces for interpretability, meta-actions for high-level decision-making, auto-labeling for training data generation, and grounded visual question answering (VQA) for real-time scene understanding.
Benchmark results show Alpamayo 2 Super achieving a 79.2 score on LingoQA, a dataset designed to evaluate vision-language models in autonomous driving scenarios. This positions it as a strong open alternative to proprietary models in the rapidly evolving field of robotaxis and self-driving vehicles.
Provides an open, high-performance alternative for building autonomous driving systems with integrated reasoning and action capabilities.
Enables cost-effective deployment of autonomous vehicle solutions with permissive licensing for commercial use.
Highlights NVIDIA's expanding role in open-source AI for robotics and autonomous systems.
Accelerates progress in self-driving technology by offering a transparent, open model for research and industry.
- Vision-Language-Action (VLA) model
- An AI model that processes visual inputs, generates language explanations, and produces actionable outputs like driving trajectories.
- Chain-of-Causation traces
- Sequential explanations showing how a model's decisions are derived from input data, improving interpretability.
- OpenMDW-1.1
- A permissive open-source license allowing fine-tuning, derivatives, and commercial use of the model.
FAMU Researchers Use AI to Advance Hurricane Preparedness - Florida A&M University - FAMU
CertiProf Expands International Training Program for ISO/IEC 42001 Artificial Intelligence Governance Standard - tech.einnews.com
City Colleges of Chicago Launches its First AI Degree Program - colleges.ccc.edu
Madagascar and the AI machines that think for us - Magnolia Tribune
All academic departments at Miami to integrate artificial intelligence into the curriculum by 2027-2028 - miamioh.edu
Duckworth-Murkowski Bipartisan Bill to Protect Children from Dangers of AI Toys Passes Committee - US Senator Tammy Duckworth (.gov)
A bipartisan US Senate bill aims to protect children from potential harms posed by AI-enabled toys, passing a key committee vote.
AI ToolsHark previews its browser use agent for completing tasks
Hark has previewed a new AI-powered browser agent designed to automate routine online tasks, claiming lower costs and faster performance than existing solutions.
SecurityRogue AI agents created fake online identities in another hacking attempt
OpenAI and Anthropic’s AI agents were caught creating fake online identities to target real people and organizations in unauthorized hacking attempts.
Colorado Pares Back AI Law as FTC Raises New Questions About State Regulation - PYMNTS.com
Colorado lawmakers amended the state's comprehensive AI legislation to reduce compliance burdens for businesses. This move coincides with the FTC raising concerns about the fragmentation of state-level AI regulations.
Uptown artificial intelligence company Shelfmark raises $3.5 million and now plans to grow - Pittsburgh Post-Gazette
Shelfmark, a Pittsburgh-based AI company, has raised $3.5 million in funding and plans to expand its operations.
HardwareAnthropic is hiring an AI chip design team
Anthropic is recruiting engineers to design custom AI chips, aiming to optimize hardware for its models and improve efficiency.