NVIDIA Alpamayo

Open VLA models, frameworks, and datasets for reasoning-based autonomous vehicles.

Overview
Demo
Benefits
Technology
Models
Getting Started
Resources
Next Steps

Overview
Demo
Benefits
Technology
Models
Getting Started
Resources
Next Steps

Overview

What Is NVIDIA Alpamayo?

The NVIDIA Alpamayo™ family of open VLA models, simulation frameworks, and physical AI datasets is helping drive the next generation of reasoning-based robotaxis and autonomous vehicles. Built on NVIDIA Cosmos™, it reasons through rare, long-tail driving scenarios and generates Chain of Causation traces that make each decision interpretable and auditable.

Alpamayo moves beyond traditional rules-based software, perceiving, reasoning, and acting within unpredictable, real-world environments. It uses human-like judgment to deliver the foundation needed for safe, scalable Level 4 autonomy.

Alpamayo 2 Super, The Frontier Open Model for Robotaxis and Autonomous Vehicles, Now Available

The frontier open reasoning model for autonomous vehicle and robotaxi development is here — commercially licensed and ready to build on.

Read the Blog Watch the video

Generate Trajectories, Reasoning Traces, and Auto-Labels with NVIDIA Alpamayo 2 Super

This blog covers how AV developers can generate trajectories, causal reasoning traces, Meta-Actions, scene Q&A, and auto-labels across their own driving data.

Read the Blog

Demo

Building Autonomous Vehicles That Reason

NVIDIA Alpamayo Portfolio of Open Models for AV Development

This demo gives voice to Alpamayo’s internal reasoning — showing what it could sound like if you could listen in, while the driver simply enjoys the ride.

Watch the Video

Benefits

What Does Alpamayo Bring to Autonomous Vehicle Development?

Open and Transparent AI

Alpamayo delivers fully open models, simulation frameworks, and datasets that promote transparency, reproducibility, and trust. This means you can inspect, extend, and fine-tune the technology to meet regional safety standards and regulatory requirements.

Reasoning-Based Autonomy

Unlike perception-only systems, Alpamayo empowers vehicles to explain why they act. Vision-language-action models reason over complex driving scenes, verbalize decision logic, and support interpretable, auditable autonomy.

High-Fidelity Simulation at Scale

With open simulation frameworks and neural reconstruction tools, Alpamayo enables closed-loop testing across diverse traffic scenarios, weather conditions, and edge cases. This greatly reduces real-world testing time and accelerates validation.

An Accelerated Path to Level 4

By standardizing foundational autonomy components, Alpamayo helps OEMs, Tier 1s, startups, and researchers shorten development cycles, align with global safety expectations, and focus on differentiation rather than rebuilding core AI systems.

Technology

Accelerated Autonomous Vehicle Development

See how Alpamayo is used across the AV development process.

Data and Labeling

Physical AI open datasets provide real-world driving data, and CoC Auto-Labeling Pipeline generates Chain-of-Causation reasoning labels from driving clips at scale.

Get Started

Model Training and
Post-Training

Alpamayo open VLA models are trained and optimized with NVIDIA AlpaGym through closed-loop RL.

Get Started

Simulation and Validation

AlpaSim validates model behavior in closed-loop simulation before real-world deployment.

Get Started

In-Vehicle Deployment

Alpamayo enables a validated path to in-vehicle deployment on NVIDIA DRIVE AGX Thor™.

Learn More

Models and Frameworks

Core Components of the Alpamayo Portfolio

Alpamayo : Open Reasoning VLA Model

Alpamayo is built on NVIDIA Cosmos™ and processes multi-camera video, navigation inputs, and driving context to generate trajectories and Chain-of-Causation reasoning traces. It’s available in 10 billion parameters with Alpamayo 1 Nano and 1.5 Nano, and scales to 34 billion parameters with Alpamayo 2 Super.

Get Started on GitHub Download on Hugging Face

AlpaSim: Open Simulation Framework

An open-source closed-loop AV simulation framework, AlpaSim validates model decisions against real-world consequences, accelerating safety validation without scaling physical test fleets.

Get Started on GitHub

AlpaGym: Closed-Loop Reinforcement Learning

NVIDIA AlpaGym is a high-throughput closed-loop RL training framework, training AV models through continuous decision and observation cycles, producing models validated against the full complexity of real-world driving.

Read the How-to Guide

Physical AI Open Datasets: Real-World Driving Data

The largest open multi-sensor driving dataset available, Physical AI Open Datasets provide real-world driving data with Chain-of-Causation reasoning labels across 25 countries, purpose-built for long-tail scenario coverage.

Download on Hugging Face Developer Kit on GitHub

Starting Options

Getting Started With Alpamayo

Explore the Open Models

Start working with the Alpamayo 2 Super reasoning foundation model.

Download the model from GitHub and Hugging Face.
Generate trajectories and reasoning traces.
Build on open weights and Apache 2.0 code.

Start Developing

Post-Train and Optimize

Adapt Alpamayo to your own driving data and compress it for the edge.

Fine-tune it on your own driving data.
Post-train it with reinforcement learning.
Quantize the model for in-vehicle inference.

Get Recipes

Get in Touch

Talk to an NVIDIA AV specialist about moving from open model to in-vehicle deployment.

Deploy in-vehicle inference on DRIVE AGX Thor.
Plan your safety and certification path.
License Alpamayo for commercial production use.

Contact Sales

Resources

Explore the Latest in Automotive

Blogs
Videos
Sessions

View All Blogs

This is NVIDIA Alpamayo Thinking Out Loud

This demo gives voice to Alpamayo’s internal reasoning, showing what it could sound like if you could listen in, while the driver simply enjoys the ride.

Watch Now

In-Car Reasoning: AI That Drives, Explains Decisions, and Responds to Passengers

This video showcases four capabilities of NVIDIA Alpamayo: chain-of-thought, verbalized reasoning, natural language Q&A, and language instructions and commands.

Watch Now

How Autonomous Vehicles Learn to Reason With NVIDIA Alpamayo

See how open NVIDIA Alpamayo models, simulation frameworks, and physical AI datasets accelerate reasoning-based development for autonomous vehicles and self-driving cars.

Watch Now

View All Sessions

Next Steps

Ready to Get Started?

Use NVIDIA Alpamayo open models, frameworks, and datasets to develop reasoning-based autonomous driving.

Start Developing

Request a Consultation

Talk to an NVIDIA AV specialist about taking reasoning-based autonomous driving from Alpamayo models into production on NVIDIA DRIVE™.

Let’s Talk

Stay Informed

Get the latest news on NVIDIA automotive technologies, breakthroughs, and more.

Contact Us

This form is for automotive inquires only. For other inquiries, please click here.

Section

Section

First Name

Last Name

Business Email Address

Business Phone Number

Job Title

Organization / University Name

Interest Area

Website

Is Your Company a Startup?

Location

Address

City

State

State/Province

Postal/Zip Code

Industry

Preferred Language

Please provide a brief description of your product and business model.

Where do you stand within your product development? Please include the list of NVIDIA products you are using.

How can NVIDIA help you achieve your goals?

Send me the latest enterprise news, announcements, and more from NVIDIA. I can unsubscribe at any time.

NVIDIA Privacy Policy

I agree to the collection and processing of the above information by NVIDIA <span class="corporation-txt hidden">Corporation </span>for the purposes of research and event organization, and I have read and agree to <a href="https://www.nvidia.com/en-us/about-nvidia/privacy-policy/?deeplink=visiting-our-website" target="_blank">NVIDIA Privacy Policy</a>.

I agree that the above information will be transferred to NVIDIA Corporation in the United States and stored in a manner consistent with <a href="https://www.nvidia.com/en-us/about-nvidia/privacy-policy/?deeplink=visiting-our-website" target="_blank">NVIDIA Privacy Policy</a> due to necessities for research, event organization and corresponding NVIDIA internal management and system operation need. You may contact us by sending an email to <a href="mailto:privacy@nvidia.com">privacy@nvidia.com</a> to resolve related problems.

NVIDIA Alpamayo

What Is NVIDIA Alpamayo?

Alpamayo 2 Super, The Frontier Open Model for Robotaxis and Autonomous Vehicles, Now Available

Generate Trajectories, Reasoning Traces, and Auto-Labels with NVIDIA Alpamayo 2 Super

Building Autonomous Vehicles That Reason

What Does Alpamayo Bring to Autonomous Vehicle Development?

Open and Transparent AI

Reasoning-Based Autonomy

High-Fidelity Simulation at Scale

An Accelerated Path to Level 4

Technology

Accelerated Autonomous Vehicle Development

Data and Labeling

Model Training and Post-Training

Simulation and Validation

In-Vehicle Deployment

Core Components of the Alpamayo Portfolio

Alpamayo : Open Reasoning VLA Model

AlpaSim: Open Simulation Framework

AlpaGym: Closed-Loop Reinforcement Learning

Physical AI Open Datasets: Real-World Driving Data

Starting Options

Getting Started With Alpamayo

Explore the Open Models

Post-Train and Optimize

Get in Touch

Explore the Latest in Automotive

This is NVIDIA Alpamayo Thinking Out Loud

In-Car Reasoning: AI That Drives, Explains Decisions, and Responds to Passengers

How Autonomous Vehicles Learn to Reason With NVIDIA Alpamayo

Next Steps

Ready to Get Started?

Request a Consultation

Stay Informed

Contact Us

Model Training and
Post-Training