---
format: "aidr-story-markdown/v1"
id: "fbdb5550e0a6a1d5aa6fbfd851de87d7d1a47ac6369b659f255d25e97b21d59e"
canonical_url: "https://aidr.today/fbdb5550?lang=en"
title: "Running AV Reasoning Models Within Onboard Compute and Latency Limits"
lang: "en"
requested_lang: "en"
available_langs: ["en","vi"]
translation_fallback: null
fallback_fields: []
published_at: "2026-11-10T16:00:00.000Z"
category: "Models"
topics: ["nvidia","reasoning"]
source_urls: ["https://developer.nvidia.com/blog/running-av-reasoning-models-within-onboard-compute-and-latency-limits/"]
summary: "NVIDIA Alpamayo 2 Edge is an adaptive reasoning vision-language-action model designed for real-time operation on NVIDIA DRIVE AGX Thor. It is built on the 2B Nemotron-based reasoning backbone of NVIDIA Cosmos 3 Edge and paired with a diffusion action expert. Alpamayo 2 Edge was post-trained with closed-loop reinforcement learning in NVIDIA AlpaGym, improving closed-loop driving performance by 48.9%."
---

# Running AV Reasoning Models Within Onboard Compute and Latency Limits

> [Open the canonical story](<https://aidr.today/fbdb5550?lang=en>)

**Published:** 2026-11-10T16:00:00.000Z
**Category:** Models
**Topics:** nvidia, reasoning

## Summary

NVIDIA Alpamayo 2 Edge is an adaptive reasoning vision\-language\-action model designed for real\-time operation on NVIDIA DRIVE AGX Thor\. It is built on the 2B Nemotron\-based reasoning backbone of NVIDIA Cosmos 3 Edge and paired with a diffusion action expert\. Alpamayo 2 Edge was post\-trained with closed\-loop reinforcement learning in NVIDIA AlpaGym, improving closed\-loop driving performance by 48\.9%\.

## Sources

- [Story source](<https://developer.nvidia.com/blog/running-av-reasoning-models-within-onboard-compute-and-latency-limits/>)