--- title: Productive, Durable, Fungible: How NVIDIA AI Factories... description: NVIDIA argues that AI factory return on investment rests on three pillars: productivity (tokens per megawatt and cost per token), durability (useful life of... canonical: https://daily.dev/posts/productive-durable-fungible-how-nvidia-ai-factories-maximize-return-on-investment-hboqzc4w5 twitter:card: summary_large_image twitter:site: @dailydotdev og:type: website og:site_name: daily.dev og:title: Productive, Durable, Fungible: How NVIDIA AI Factories Maximize Return on Investment | daily.dev og:description: NVIDIA argues that AI factory return on investment rests on three pillars: productivity (tokens per megawatt and cost per token), durability (useful life of... og:url: https://daily.dev/posts/productive-durable-fungible-how-nvidia-ai-factories-maximize-return-on-investment-hboqzc4w5 og:image: https://api.daily.dev/og/posts/HbOQzc4w5.png og:image:alt: Productive, Durable, Fungible: How NVIDIA AI Factories Maximize Return on Investment 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. # Productive, Durable, Fungible: How NVIDIA AI Factories Maximize Return on Investment **[NVIDIA](https://daily.dev/sources/nvidia)** · 7 min read · 0 upvotes · 0 comments ## Summary NVIDIA argues that AI factory return on investment rests on three pillars: productivity (tokens per megawatt and cost per token), durability (useful life of installed GPUs), and fungibility (ability to run many workload types). It cites SemiAnalysis data claiming Vera Rubin NVL72 delivers over 30x higher throughput per megawatt and up to 45x lower cost per million tokens versus GB300 NVL72 on DeepSeek V4 Pro. It also highlights that A100 GPUs shipped in 2020 remain commercially useful, with CoreWeave extending bookings through 2029, and cites third-party analyses on depreciation schedules and resale value. Customer examples (Lilly, Pinterest, Revolut, Runway, Texas A&M, Dassault Systèmes, Unilever) illustrate workload diversity across AI and non-AI use cases, all framed as maximizing ROI for AI factory operators. ## Full article daily.dev links to this article rather than hosting it. Read it at the original source: ## Questions this post answers ### How much faster is NVIDIA Vera Rubin NVL72 compared to GB300 NVL72 for inference workloads? Vera Rubin NVL72 delivers over 30 times higher throughput per megawatt and up to 45 times lower cost per million tokens than GB300 NVL72 on the DeepSeek V4 Pro model, according to SemiAnalysis AgentX benchmark data. These gains come from codesign across models, software, compute, networking, and memory rather than any single component improvement. _Teams weighing GPU generation upgrades can track benchmark comparisons like this one on daily.dev._ ### How long do NVIDIA A100 GPUs remain economically useful in data centers? A100 GPUs shipped in 2020 are still in commercial service six years later, with CoreWeave extending bookings for original 2020 units through 2029. Silicon Data reports a six-year-old A100 still retains roughly a quarter of its original value, well past where a five-year depreciation schedule would have placed it at zero. _Anyone planning long-term GPU procurement can follow hardware lifespan data on daily.dev._ ## Similar posts on daily.dev - [Understanding the economics of AI factories](https://daily.dev/posts/understanding-the-economics-of-ai-factories-5c6ecwvmj) · InfoWorld · 0 upvotes · 0 comments - [Scaling Token Factory Revenue and AI Efficiency by Maximizing Performance per Watt](https://daily.dev/posts/scaling-token-factory-revenue-and-ai-efficiency-by-maximizing-performance-per-watt-2qzgn7uql) · NVIDIA Developer · 0 upvotes · 0 comments - [AI Factories: The New Infrastructure of Intelligence](https://daily.dev/posts/ai-factories-the-new-infrastructure-of-intelligence-vzxgrxqdx) · NVIDIA · 0 upvotes · 0 comments - [Efficiency at Scale: NVIDIA, Energy Leaders Accelerating Power‑Flexible AI Factories to Fortify the Grid](https://daily.dev/posts/efficiency-at-scale-nvidia-energy-leaders-accelerating-power-flexible-ai-factories-to-fortify-the--ihtvnvz4p) · NVIDIA · 1 upvotes · 0 comments - [NVIDIA Vera Rubin and Blackwell Set a New Standard for Agentic AI Performance per Watt](https://daily.dev/posts/nvidia-vera-rubin-and-blackwell-set-a-new-standard-for-agentic-ai-performance-per-watt-r1iqkqtkn) · NVIDIA Developer · 0 upvotes · 0 comments --- Tags: [#nvidia](https://daily.dev/tags/nvidia), [#gpu](https://daily.dev/tags/gpu), [#ai-infrastructure](https://daily.dev/tags/ai-infrastructure), [#cuda](https://daily.dev/tags/cuda) [View this post on daily.dev](https://daily.dev/posts/productive-durable-fungible-how-nvidia-ai-factories-maximize-return-on-investment-hboqzc4w5) ```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":"Productive, Durable, Fungible: How NVIDIA AI Factories Maximize Return on Investment","url":"https://daily.dev/posts/productive-durable-fungible-how-nvidia-ai-factories-maximize-return-on-investment-hboqzc4w5","mainEntityOfPage":{"@type":"WebPage","@id":"https://daily.dev/posts/productive-durable-fungible-how-nvidia-ai-factories-maximize-return-on-investment-hboqzc4w5"},"datePublished":"2026-10-01T14:37:14.407Z","dateModified":"2026-10-01T14:37:36.605Z","description":"NVIDIA argues that AI factory return on investment rests on three pillars: productivity (tokens per megawatt and cost per token), durability (useful life of...","image":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/153b4b8eae834439c769776e3b2e0587?_a=AQAEuop","thumbnailUrl":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/153b4b8eae834439c769776e3b2e0587?_a=AQAEuop","isAccessibleForFree":true,"articleSection":"NVIDIA","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":"NVIDIA","logo":"https://media.daily.dev/image/upload/t_logo,f_auto/v1/logos/236a5c1f410e450b8d696cf3a4bbe374","url":"https://daily.dev/sources/nvidia"},"commentCount":0,"discussionUrl":"https://daily.dev/posts/productive-durable-fungible-how-nvidia-ai-factories-maximize-return-on-investment-hboqzc4w5","interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":0},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":0}],"keywords":"nvidia,gpu,ai-infrastructure,cuda","timeRequired":"PT7M"} {"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://daily.dev"},{"@type":"ListItem","position":2,"name":"NVIDIA","item":"https://daily.dev/sources/nvidia"},{"@type":"ListItem","position":3,"name":"Productive, Durable, Fungible: How NVIDIA AI Factories Maximize Return on Investment"}]} {"@context":"https://schema.org","@type":"FAQPage","@id":"https://daily.dev/posts/productive-durable-fungible-how-nvidia-ai-factories-maximize-return-on-investment-hboqzc4w5#faq","mainEntity":[{"@type":"Question","name":"How much faster is NVIDIA Vera Rubin NVL72 compared to GB300 NVL72 for inference workloads?","acceptedAnswer":{"@type":"Answer","text":"Vera Rubin NVL72 delivers over 30 times higher throughput per megawatt and up to 45 times lower cost per million tokens than GB300 NVL72 on the DeepSeek V4 Pro model, according to SemiAnalysis AgentX benchmark data. These gains come from codesign across models, software, compute, networking, and memory rather than any single component improvement. Teams weighing GPU generation upgrades can track benchmark comparisons like this one on daily.dev."}},{"@type":"Question","name":"How long do NVIDIA A100 GPUs remain economically useful in data centers?","acceptedAnswer":{"@type":"Answer","text":"A100 GPUs shipped in 2020 are still in commercial service six years later, with CoreWeave extending bookings for original 2020 units through 2029. Silicon Data reports a six-year-old A100 still retains roughly a quarter of its original value, well past where a five-year depreciation schedule would have placed it at zero. Anyone planning long-term GPU procurement can follow hardware lifespan data on daily.dev."}}]} ```