--- title: Compute that takes any shape | daily.dev description: Vercel details Fluid, its unified compute architecture that dynamically assembles machines to fit any workload, builds, functions, or sandboxes, instead of... canonical: https://daily.dev/posts/compute-that-takes-any-shape-8fxq1lmnr twitter:card: summary_large_image twitter:site: @dailydotdev og:type: website og:site_name: daily.dev og:title: Compute that takes any shape | daily.dev og:description: Vercel details Fluid, its unified compute architecture that dynamically assembles machines to fit any workload, builds, functions, or sandboxes, instead of... og:url: https://daily.dev/posts/compute-that-takes-any-shape-8fxq1lmnr og:image: https://api.daily.dev/og/posts/8Fxq1LMNR.png og:image:alt: Compute that takes any shape 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. # Compute that takes any shape **[Vercel](https://daily.dev/sources/vercel)** · 5 min read · 2 upvotes · 0 comments ## Summary Vercel details Fluid, its unified compute architecture that dynamically assembles machines to fit any workload, builds, functions, or sandboxes, instead of requiring separate infrastructure for each. Three components make it work: Hive (the hardware layer provisioning isolated VMs at global scale), Fluid images (a custom boot format called VHS that lets users run their own OS images and boot in milliseconds), and Vercel Drives (portable storage decoupled from compute, currently in private beta for sandboxes). Vercel reports the system now handles over 15 million builds a day, 25 million sandboxes a week, and a trillion requests a month, positioning it as especially suited to the fast provisioning agents need. ## Full article daily.dev links to this article rather than hosting it. Read it at the original source: ## Questions this post answers ### What is Vercel Fluid compute and how does it work? Fluid is Vercel's unified compute architecture that assembles a machine on demand to fit whatever workload arrives, builds, functions, sandboxes, or agent tasks. It relies on three parts: Hive, which provisions isolated VMs (often already warm) in milliseconds; Fluid images, a boot format called VHS that lets custom OS images resume instead of boot; and Vercel Drives, portable storage decoupled from the machine. _Engineers weighing serverless platforms for agent workloads can track infrastructure shifts like this via daily.dev._ ### What is Vercel's VHS (Vercel Hive Snapshot) format used for? VHS is the optimized boot format Vercel uses to convert container images into a snapshot that can resume rather than boot from scratch, making a custom machine ready in milliseconds. It underlies both Dockerfile deploys and sandbox custom images, and the same technology powers Sandbox Snapshots, so improvements apply across products. Vercel's own v0 product already uses it for its development environments. _Developers evaluating fast-boot sandboxing for agents can follow these infrastructure details on daily.dev._ ### What is Vercel Drives and when will it be generally available? Vercel Drives is portable, durable storage that travels with a workload rather than being tied to one machine's disk, letting compute underneath be swapped while state persists across sessions. It attaches to sandboxes today in private beta, with plans to extend it to the rest of the Fluid compute system, including functions and builds, over time. _Teams designing stateful agent or sandbox workflows can stay on top of storage changes like this through daily.dev._ --- Tags: [#cloud](https://daily.dev/tags/cloud), [#ai-agents](https://daily.dev/tags/ai-agents), [#containers](https://daily.dev/tags/containers), [#serverless](https://daily.dev/tags/serverless), [#vercel](https://daily.dev/tags/vercel) [View this post on daily.dev](https://daily.dev/posts/compute-that-takes-any-shape-8fxq1lmnr) ```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":"Compute that takes any shape","url":"https://daily.dev/posts/compute-that-takes-any-shape-8fxq1lmnr","mainEntityOfPage":{"@type":"WebPage","@id":"https://daily.dev/posts/compute-that-takes-any-shape-8fxq1lmnr"},"datePublished":"2026-09-01T16:31:38.053Z","dateModified":"2026-09-01T16:42:05.403Z","description":"Vercel details Fluid, its unified compute architecture that dynamically assembles machines to fit any workload, builds, functions, or sandboxes, instead of...","image":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/29c265482bcc782be8aef1771bc5fa55?_a=AQAEuop","thumbnailUrl":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/29c265482bcc782be8aef1771bc5fa55?_a=AQAEuop","isAccessibleForFree":true,"articleSection":"Vercel","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":"Vercel","logo":"https://media.daily.dev/image/upload/t_logo,f_auto/v1/logos/04cca7235cf84db092f24b7ba427b3e1","url":"https://daily.dev/sources/vercel"},"commentCount":0,"discussionUrl":"https://daily.dev/posts/compute-that-takes-any-shape-8fxq1lmnr","interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":2},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":0}],"keywords":"cloud,ai-agents,containers,serverless,vercel","timeRequired":"PT5M"} {"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://daily.dev"},{"@type":"ListItem","position":2,"name":"Vercel","item":"https://daily.dev/sources/vercel"},{"@type":"ListItem","position":3,"name":"Compute that takes any shape"}]} {"@context":"https://schema.org","@type":"FAQPage","@id":"https://daily.dev/posts/compute-that-takes-any-shape-8fxq1lmnr#faq","mainEntity":[{"@type":"Question","name":"What is Vercel Fluid compute and how does it work?","acceptedAnswer":{"@type":"Answer","text":"Fluid is Vercel's unified compute architecture that assembles a machine on demand to fit whatever workload arrives, builds, functions, sandboxes, or agent tasks. It relies on three parts: Hive, which provisions isolated VMs (often already warm) in milliseconds; Fluid images, a boot format called VHS that lets custom OS images resume instead of boot; and Vercel Drives, portable storage decoupled from the machine. Engineers weighing serverless platforms for agent workloads can track infrastructure shifts like this via daily.dev."}},{"@type":"Question","name":"What is Vercel's VHS (Vercel Hive Snapshot) format used for?","acceptedAnswer":{"@type":"Answer","text":"VHS is the optimized boot format Vercel uses to convert container images into a snapshot that can resume rather than boot from scratch, making a custom machine ready in milliseconds. It underlies both Dockerfile deploys and sandbox custom images, and the same technology powers Sandbox Snapshots, so improvements apply across products. Vercel's own v0 product already uses it for its development environments. Developers evaluating fast-boot sandboxing for agents can follow these infrastructure details on daily.dev."}},{"@type":"Question","name":"What is Vercel Drives and when will it be generally available?","acceptedAnswer":{"@type":"Answer","text":"Vercel Drives is portable, durable storage that travels with a workload rather than being tied to one machine's disk, letting compute underneath be swapped while state persists across sessions. It attaches to sandboxes today in private beta, with plans to extend it to the rest of the Fluid compute system, including functions and builds, over time. Teams designing stateful agent or sandbox workflows can stay on top of storage changes like this through daily.dev."}}]} ```