--- title: Cosmos3-Edge, Cosmos3-Nano, and Cosmos3-Super models now... description: NVIDIA's Cosmos 3 family of open, frontier omnimodal world models for physical AI is now available on Amazon SageMaker JumpStart. The family includes... canonical: https://daily.dev/posts/cosmos3-edge-cosmos3-nano-and-cosmos3-super-models-now-available-on-amazon-sagemaker-jumpstart-4ec9r48zh twitter:card: summary_large_image twitter:site: @dailydotdev og:type: website og:site_name: daily.dev og:title: Cosmos3-Edge, Cosmos3-Nano, and Cosmos3-Super models now available on Amazon SageMaker JumpStart | daily.dev og:description: NVIDIA's Cosmos 3 family of open, frontier omnimodal world models for physical AI is now available on Amazon SageMaker JumpStart. The family includes... og:url: https://daily.dev/posts/cosmos3-edge-cosmos3-nano-and-cosmos3-super-models-now-available-on-amazon-sagemaker-jumpstart-4ec9r48zh og:image: https://api.daily.dev/og/posts/4Ec9r48ZH.png og:image:alt: Cosmos3-Edge, Cosmos3-Nano, and Cosmos3-Super models now available on Amazon SageMaker JumpStart 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. # Cosmos3-Edge, Cosmos3-Nano, and Cosmos3-Super models now available on Amazon SageMaker JumpStart **[AWS](https://daily.dev/sources/aws)** · 2 min read · 1 upvotes · 1 comments ## Summary NVIDIA's Cosmos 3 family of open, frontier omnimodal world models for physical AI is now available on Amazon SageMaker JumpStart. The family includes Cosmos3-Edge (4B parameters, for on-device robot control and real-time reasoning on Jetson Thor edge hardware), Cosmos3-Nano (16B parameters, physics-aware world generation supporting text/image/video/audio inputs), and Cosmos3-Super (64B parameters, highest-fidelity world generation and simulation using a Mixture-of-Transformers architecture). Customers can deploy any of these models with a few clicks via the SageMaker JumpStart console or Python SDK for use cases like robotics, autonomous vehicles, and vision AI. ## Full article daily.dev links to this article rather than hosting it. Read it at the original source: ## Questions this post answers ### What are the parameter sizes and specs of the NVIDIA Cosmos3-Edge, Cosmos3-Nano, and Cosmos3-Super models? Cosmos3-Edge is a 4B-parameter omni-model with a 2B Nemotron-based reasoner, running at 640x360 robot-control resolution and generating 32 actions per inference at 15 Hz on NVIDIA Jetson Thor, supporting 256p/480p video at 12-30 FPS. Cosmos3-Nano is 16B parameters, handling text, image, video, audio, and action trajectories up to 720p with chain-of-thought reasoning. Cosmos3-Super is 64B parameters, using a Mixture-of-Transformers architecture for language, image, video, audio, and action generation up to 720p. _Teams choosing a physical AI model size can track Cosmos 3 availability updates on daily.dev._ ### How do I deploy NVIDIA Cosmos 3 models on AWS SageMaker? Cosmos3-Edge, Cosmos3-Nano, and Cosmos3-Super can be deployed with a few clicks through the SageMaker JumpStart model catalog in the SageMaker console, or programmatically using the SageMaker Python SDK to deploy directly to an AWS account. Full deployment guidance is available in the Amazon SageMaker JumpStart documentation. _daily.dev helps developers stay on top of new foundation model deployment options as they ship._ ## Community discussion Top comments from developers on daily.dev. **@trevorsuna** · 0 upvotes > The family split is practical: on-device control, mid-sized world generation, and high-fidelity simulation have very different latency and cost envelopes. The important benchmark will be end-to-end robotics loops, including sensor transfer and deployment startup, rather than model quality in isolation. --- Tags: [#ai](https://daily.dev/tags/ai), [#aws](https://daily.dev/tags/aws), [#nvidia](https://daily.dev/tags/nvidia), [#robotics](https://daily.dev/tags/robotics) [View this post on daily.dev](https://daily.dev/posts/cosmos3-edge-cosmos3-nano-and-cosmos3-super-models-now-available-on-amazon-sagemaker-jumpstart-4ec9r48zh) ```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":"Cosmos3-Edge, Cosmos3-Nano, and Cosmos3-Super models now available on Amazon SageMaker JumpStart","url":"https://daily.dev/posts/cosmos3-edge-cosmos3-nano-and-cosmos3-super-models-now-available-on-amazon-sagemaker-jumpstart-4ec9r48zh","mainEntityOfPage":{"@type":"WebPage","@id":"https://daily.dev/posts/cosmos3-edge-cosmos3-nano-and-cosmos3-super-models-now-available-on-amazon-sagemaker-jumpstart-4ec9r48zh"},"datePublished":"2026-08-28T00:07:52.793Z","dateModified":"2026-08-28T02:01:54.076Z","description":"NVIDIA's Cosmos 3 family of open, frontier omnimodal world models for physical AI is now available on Amazon SageMaker JumpStart. The family includes...","image":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/e82c8cceed8771f89f06dfc50fa9305d?_a=AQAEuop","thumbnailUrl":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/e82c8cceed8771f89f06dfc50fa9305d?_a=AQAEuop","isAccessibleForFree":true,"articleSection":"AWS","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":"AWS","logo":"https://media.daily.dev/image/upload/t_logo,f_auto/v1/logos/aws","url":"https://daily.dev/sources/aws"},"commentCount":1,"discussionUrl":"https://daily.dev/posts/cosmos3-edge-cosmos3-nano-and-cosmos3-super-models-now-available-on-amazon-sagemaker-jumpstart-4ec9r48zh","interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":1},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":1}],"keywords":"ai,aws,nvidia,robotics","timeRequired":"PT2M"} {"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://daily.dev"},{"@type":"ListItem","position":2,"name":"AWS","item":"https://daily.dev/sources/aws"},{"@type":"ListItem","position":3,"name":"Cosmos3-Edge, Cosmos3-Nano, and Cosmos3-Super models now available on Amazon SageMaker JumpStart"}]} {"@context":"https://schema.org","@type":"WebPage","@id":"https://daily.dev/posts/cosmos3-edge-cosmos3-nano-and-cosmos3-super-models-now-available-on-amazon-sagemaker-jumpstart-4ec9r48zh","comment":[{"@type":"Comment","text":"The family split is practical: on-device control, mid-sized world generation, and high-fidelity simulation have very different latency and cost envelopes. The important benchmark will be end-to-end robotics loops, including sensor transfer and deployment startup, rather than model quality in isolation.","datePublished":"2026-08-28T13:47:43.153Z","url":"https://daily.dev/posts/4Ec9r48ZH#c-ypgdgAoEj","author":{"@type":"Person","name":"Trevor Suna","url":"https://daily.dev/trevorsuna","image":"https://media.daily.dev/image/upload/s--dZ7gXxpp--/f_auto/v1784081551/avatars/avatar_EMoP47rpuw8DNjhp6R1b6?_a=BAMAMicg0"}}]} {"@context":"https://schema.org","@type":"FAQPage","@id":"https://daily.dev/posts/cosmos3-edge-cosmos3-nano-and-cosmos3-super-models-now-available-on-amazon-sagemaker-jumpstart-4ec9r48zh#faq","mainEntity":[{"@type":"Question","name":"What are the parameter sizes and specs of the NVIDIA Cosmos3-Edge, Cosmos3-Nano, and Cosmos3-Super models?","acceptedAnswer":{"@type":"Answer","text":"Cosmos3-Edge is a 4B-parameter omni-model with a 2B Nemotron-based reasoner, running at 640x360 robot-control resolution and generating 32 actions per inference at 15 Hz on NVIDIA Jetson Thor, supporting 256p/480p video at 12-30 FPS. Cosmos3-Nano is 16B parameters, handling text, image, video, audio, and action trajectories up to 720p with chain-of-thought reasoning. Cosmos3-Super is 64B parameters, using a Mixture-of-Transformers architecture for language, image, video, audio, and action generation up to 720p. Teams choosing a physical AI model size can track Cosmos 3 availability updates on daily.dev."}},{"@type":"Question","name":"How do I deploy NVIDIA Cosmos 3 models on AWS SageMaker?","acceptedAnswer":{"@type":"Answer","text":"Cosmos3-Edge, Cosmos3-Nano, and Cosmos3-Super can be deployed with a few clicks through the SageMaker JumpStart model catalog in the SageMaker console, or programmatically using the SageMaker Python SDK to deploy directly to an AWS account. Full deployment guidance is available in the Amazon SageMaker JumpStart documentation. daily.dev helps developers stay on top of new foundation model deployment options as they ship."}}]} ```