--- title: NVIDIA Alpamayo 2 Super: an open reasoning model for... description: NVIDIA has released Alpamayo 2 Super, a 34-billion-parameter open vision-language-action model designed for autonomous vehicle development. It combines a 32B... canonical: https://daily.dev/posts/nvidia-alpamayo-2-super-an-open-reasoning-model-for-autonomous-vehicles-now-available-commercially-rn5mgmzgy twitter:card: summary_large_image twitter:site: @dailydotdev og:type: website og:site_name: daily.dev og:title: NVIDIA Alpamayo 2 Super: an open reasoning model for autonomous vehicles, now available commercially | daily.dev og:description: NVIDIA has released Alpamayo 2 Super, a 34-billion-parameter open vision-language-action model designed for autonomous vehicle development. It combines a 32B... og:url: https://daily.dev/posts/nvidia-alpamayo-2-super-an-open-reasoning-model-for-autonomous-vehicles-now-available-commercially-rn5mgmzgy og:image: https://api.daily.dev/og/posts/Rn5mgmzgy.png og:image:alt: NVIDIA Alpamayo 2 Super: an open reasoning model for autonomous vehicles, now available commercially og:image:width: 1200 og:image:height: 630 og:locale: en --- > ## Documentation Index > Fetch the complete documentation index at: https://daily.dev/llms.txt > Use this file to discover all available pages before exploring further. # NVIDIA Alpamayo 2 Super: an open reasoning model for autonomous vehicles, now available commercially **[Collections](https://daily.dev/sources/collections)** · 2 min read · 0 upvotes · 0 comments ## Summary NVIDIA has released Alpamayo 2 Super, a 34-billion-parameter open vision-language-action model designed for autonomous vehicle development. It combines a 32B reasoning backbone (Cosmos 3 Super Reasoner) with a 2B diffusion-based Action Expert, post-trained with reinforcement learning and capable of processing 360-degree surround-view input from up to seven cameras. The model produces five coupled outputs per driving scenario: planned trajectories, Chain-of-Causation reasoning traces, high-level meta-actions, visual QA with 2D grounding, and structured auto-labels. It tops the LingoQA autonomous driving benchmark with a score of 79.2, outperforming Gemini 2.5 Pro and GPT-4o. The intended deployment pattern is cloud-to-car: run the full model in the cloud for synthetic data generation and distillation, then deploy smaller distilled models in production vehicles. It integrates with NVIDIA Halos safety-validation workflows and is designed to meet ISO/PAS 8800 requirements. Available commercially under the OpenMDW-1.1 license, with inference notebooks on GitHub and weights on Hugging Face. ## Content NVIDIA has released Alpamayo 2 Super, an open vision-language-action model built for autonomous vehicle development. It's available now for commercial use under the OpenMDW-1.1 license, which permits fine-tuning and redistribution. ## What it is Alpamayo 2 Super is a 34-billion-parameter model that combines a 32B reasoning backbone (Cosmos 3 Super Reasoner) with a 2B diffusion-based Action Expert. It was post-trained with reinforcement learning and accepts 360-degree surround-view input from up to seven cameras. For each driving scenario, the model produces five coupled outputs: - Planned ego-vehicle trajectories - Chain-of-Causation reasoning traces - High-level meta-actions (yield, lane change, stop) - Visual question answering with 2D grounding - Structured auto-labels for training data The reasoning traces are the part worth paying attention to. Rather than producing a trajectory and leaving you to guess why, the model generates an explicit causal chain explaining its decisions — which matters a lot when you're trying to validate behavior or generate synthetic training data. ## Benchmark results Alpamayo 2 Super tops the LingoQA autonomous driving benchmark with a score of 79.2, ranking first among 37 evaluated models. For comparison, Gemini 2.5 Pro and GPT-4o score significantly lower. On VQA answer similarity, it scores 0.652 versus 0.450 for Qwen3-VL 32B. Its AlpaSim closed-loop score is 1.50. ## How it fits into a production workflow The intended deployment pattern is cloud-to-car: run the full frontier-scale model in the cloud for synthetic data generation and model distillation, then deploy the distilled, smaller models in production vehicles where compute is constrained. This sidesteps the obvious problem of running a 34B model on embedded hardware. The model integrates with NVIDIA Halos safety-validation workflows and is designed to meet ISO/PAS 8800 requirements for AI in road vehicles. ## The broader ecosystem Alpamayo 2 Super sits inside a larger toolkit. AlpaSim handles closed-loop simulation, AlpaGym supports reinforcement learning training, and NVIDIA has released open datasets alongside the model. The Alpamayo model family has passed 500,000 downloads on Hugging Face. Inference notebooks are available on GitHub and weights on Hugging Face. ## Similar posts on daily.dev - [Nvidia launches Alpamayo, open AI models that allow autonomous vehicles to ‘think like a human’](https://daily.dev/posts/nvidia-launches-alpamayo-open-ai-models-that-allow-autonomous-vehicles-to-think-like-a-human--9skkzpk7x) · TechCrunch · 0 upvotes · 0 comments - [Building Autonomous Vehicles That Reason with NVIDIA Alpamayo](https://daily.dev/posts/building-autonomous-vehicles-that-reason-with-nvidia-alpamayo-busbkymhb) · NVIDIA Developer · 0 upvotes · 0 comments - [AlpamayoR1: Large Causal Reasoning Models for Autonomous Driving](https://daily.dev/posts/alpamayor1-large-causal-reasoning-models-for-autonomous-driving-xdvgot0xf) · Towards Data Science · 1 upvotes · 0 comments --- Tags: [#nvidia](https://daily.dev/tags/nvidia), [#reinforcement-learning](https://daily.dev/tags/reinforcement-learning) [View this post on daily.dev](https://daily.dev/posts/nvidia-alpamayo-2-super-an-open-reasoning-model-for-autonomous-vehicles-now-available-commercially-rn5mgmzgy) ```json {"@context":"https://schema.org","@graph":[{"@type":"Organization","@id":"https://daily.dev/#organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180},"sameAs":["https://twitter.com/dailydotdev","https://github.com/dailydotdev","https://www.linkedin.com/company/daily-dev-ltd"]},{"@type":"WebSite","@id":"https://daily.dev/#website","url":"https://daily.dev","name":"daily.dev","publisher":{"@id":"https://daily.dev/#organization"},"potentialAction":{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https://daily.dev/search?q={search_term_string}"},"query-input":"required name=search_term_string"}}]} {"@context":"https://schema.org","@type":"TechArticle","headline":"NVIDIA Alpamayo 2 Super: an open reasoning model for autonomous vehicles, now available commercially","url":"https://daily.dev/posts/nvidia-alpamayo-2-super-an-open-reasoning-model-for-autonomous-vehicles-now-available-commercially-rn5mgmzgy","mainEntityOfPage":{"@type":"WebPage","@id":"https://daily.dev/posts/nvidia-alpamayo-2-super-an-open-reasoning-model-for-autonomous-vehicles-now-available-commercially-rn5mgmzgy"},"datePublished":"2026-08-04T16:22:42.499Z","dateModified":"2026-08-04T16:23:26.621Z","description":"NVIDIA has released Alpamayo 2 Super, a 34-billion-parameter open vision-language-action model designed for autonomous vehicle development. It combines a 32B...","image":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/0daae695ea5acad463aefd6fc77017a6?_a=AQAEuop","thumbnailUrl":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/0daae695ea5acad463aefd6fc77017a6?_a=AQAEuop","isAccessibleForFree":true,"articleSection":"Collections","inLanguage":"en","publisher":{"@type":"Organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180}},"author":{"@type":"Organization","name":"Collections","logo":"https://media.daily.dev/image/upload/s--fk_6ycEi--/f_auto,q_auto/v1780996001/logos/collections?_a=BAMAMiWQ0","url":"https://daily.dev/sources/collections"},"commentCount":0,"discussionUrl":"https://daily.dev/posts/nvidia-alpamayo-2-super-an-open-reasoning-model-for-autonomous-vehicles-now-available-commercially-rn5mgmzgy","interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":0},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":0}],"keywords":"nvidia,reinforcement-learning","timeRequired":"PT2M"} {"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://daily.dev"},{"@type":"ListItem","position":2,"name":"Collections","item":"https://daily.dev/sources/collections"},{"@type":"ListItem","position":3,"name":"NVIDIA Alpamayo 2 Super: an open reasoning model for autonomous vehicles, now available commercially"}]} ```