Nvidia Unveils ‘Alpamayo’, Open-Source AI Models For Autonomous Vehicles With ‘Human-Like’ Judgement

• The Alpamayo model includes Alpamayo 1, AlpaSim and Physical AI Open Datasets. 
• “The ChatGPT moment for physical AI is here — when machines begin to understand, reason and act in the real world,” said Huang.
• Companies including Lucid, JLR, Uber and Berkeley DeepDrive are already expressing their interest in Alpamayo.

NVIDIA Corp. (NVDA) announced the launch of its ‘Alpamayo’ open AI models, simulation tools and datasets that will enable self-driving vehicles to perceive, reason and act with humanlike judgment.

At the Consumer Electronics Show (CES) 2026, Nvidia CEO Jensen Huang released the open reasoning vision language action (VLA) model including Alpamayo 1, AlpaSim and Physical AI Open Datasets.

This model will enable developers to fine-tune, distill and test models that unlock greater safety, robustness and scalability.

“The ChatGPT moment for physical AI is here — when machines begin to understand, reason and act in the real world,” said Huang in a statement.

“Robotaxis are among the first to benefit. Alpamayo brings reasoning to autonomous vehicles, allowing them to think through rare scenarios, drive safely in complex environments and explain their driving decisions — it’s the foundation for safe, scalable autonomy,” he added.

Shares of NVDA climbed 0.5% in Tuesday’s after-market hours.

What Is Alpamayo?

Alpamayo 1 is at the center of Nvidia’s open AI model. Touted to be the industry’s first chain-of-thought reasoning VLA model that is designed for the AV research community, it has a 10-billion-parameter architecture that can be used to adapt it into smaller runtime models for vehicle development. It is available on Hugging Face.

AlpaSim is the company’s fully open‑source, end-to-end simulation framework for high‑fidelity AV development that is available on GitHub. It comes with realistic sensor modeling capabilities, configurable traffic dynamics, and scalable closed‑loop testing environments that can be leveraged for rapid validation and policy refinement.

The company also announced physical AI open datasets for AV that contain 1,700+ hours of driving data across wide geographies and conditions for advancing reasoning architectures.

“Not only does [Alpamayo] take sensor input and activate steering wheel, brakes and acceleration, it also reasons about what action it is about to take,” Huang said at the CES on Monday. “It tells you what action is going to take, the reasons by which it came about that action. And then, of course, the trajectory.”

Companies including Lucid, JLR, Uber and Berkeley DeepDrive are already expressing their interest in Alpamayo to develop reasoning-based AV stacks to enable level 4 autonomy, the company said in the release.

How Did Stocktwits Users React?

On Stocktwits, retail sentiment around NVDA shares jumped to ‘bullish’ from ‘bearish’ territory a day ago amid ‘normal’ message levels at the time of writing.

Shares of NVDA are up over 25% in the past year.

For updates and corrections, email newsroom[at]stocktwits[dot]com.

Share

·

Add us on