---
title: "Generate Trajectories, Reasoning Traces, and Auto-Labels with NVIDIA Alpamayo 2 Super"
description: "Autonomous vehicle development typically relies on disjoint models for trajectory generation, intent prediction, scene understanding, and data labeling. NVIDIA Alpamayo 2 Super addresses this fragmentation with an open 34-billion-parameter vision-language-action model that couples a 32-billion-parameter Cosmos 3 Super Reasoner with a 2-billion-parameter diffusion-based Action Expert. Processing surround-view video across up to seven cameras, the model simultaneously produces future vehicle trajectories, Chain-of-Causation reasoning traces, high-level meta-actions, and grounded scene question answers. In benchmark evaluations, the model achieved a 6.4-second minADE_6 of 0.911 meters on the Physical AI AV Dataset and ranked first on LingoQA with a score of 79.2. Released under the OpenMDW-1.1 permissive license, the architecture serves as an offline policy teacher, evaluation critic, or auto-labeling engine that can be distilled for NVIDIA DRIVE AGX Thor deployment."
---

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

[NVIDIA](https://yomu.fyi/company/nvidia-developer-blog) · Elizabeth Goodman · Aug 4, 2026

**Type:** Announcement

## Summary

Autonomous vehicle development typically relies on disjoint models for trajectory generation, intent prediction, scene understanding, and data labeling. NVIDIA Alpamayo 2 Super addresses this fragmentation with an open 34-billion-parameter vision-language-action model that couples a 32-billion-parameter Cosmos 3 Super Reasoner with a 2-billion-parameter diffusion-based Action Expert. Processing surround-view video across up to seven cameras, the model simultaneously produces future vehicle trajectories, Chain-of-Causation reasoning traces, high-level meta-actions, and grounded scene question answers. In benchmark evaluations, the model achieved a 6.4-second minADE\_6 of 0.911 meters on the Physical AI AV Dataset and ranked first on LingoQA with a score of 79.2. Released under the OpenMDW-1.1 permissive license, the architecture serves as an offline policy teacher, evaluation critic, or auto-labeling engine that can be distilled for NVIDIA DRIVE AGX Thor deployment.

## Context

Autonomous vehicle development often relies on separate models for trajectory generation, intent prediction, scene understanding, and data labeling. This separation complicates comparing related outputs, investigating model behavior, and reusing representations across the development workflow.

## Approach / What changed

NVIDIA Alpamayo 2 Super integrates a 32-billion-parameter Cosmos 3 Super Reasoner with a 2-billion-parameter diffusion-based Action Expert, post-trained via reinforcement learning. The system handles 360-degree input from up to seven cameras alongside language context and motion history. It outputs future ego-vehicle trajectories, Chain-of-Causation reasoning traces, meta-actions, grounded scene question answers, and auto-labels across open-loop and closed-loop simulation environments like NVIDIA AlpaSim.

## Takeaways

- Alpamayo 2 Super combines a 32B Cosmos 3 Super Reasoner and a 2B diffusion Action Expert to jointly output future driving trajectories and Chain-of-Causation reasoning traces.
- The model achieves a 6.4-second minADE\_6 of 0.911 m on the Physical AI AV Dataset, a 0.433 AV reasoning benchmark score, and ranks first on LingoQA with a score of 79.2.
- Released under the permissive OpenMDW-1.1 license, the model can generate auto-labels with or without ground-truth future trajectories and be distilled for NVIDIA DRIVE AGX Thor hardware.

**Tags:** [LLMs](https://yomu.fyi/topic/llm), [Machine Learning](https://yomu.fyi/topic/machine-learning), [Open Source](https://yomu.fyi/topic/open-source), [Python](https://yomu.fyi/topic/python)

- Source: [NVIDIA](https://developer.nvidia.com/blog/generate-trajectories-reasoning-traces-and-auto-labels-with-nvidia-alpamayo-2-super)
- Source URL: https://developer.nvidia.com/blog/generate-trajectories-reasoning-traces-and-auto-labels-with-nvidia-alpamayo-2-super
- Ingested by Yomu: 2026-08-28T04:00:56.542Z

[Read original post](https://developer.nvidia.com/blog/generate-trajectories-reasoning-traces-and-auto-labels-with-nvidia-alpamayo-2-super)