NVIDIA’s 1 October blog post on AI-factory returns cites five outside sources. This publication opened three of the originals, SemiAnalysis, Sprout and Ornn Data, and could not open Silicon Data or Barkr. Where the originals add detail, such as an engine label or a comparison the NVIDIA post does not include, this piece reports it. NVIDIA sells the hardware it discusses, and nothing here is independently verified beyond reading those sources.
Business Pill · HOW LONG IT KEEPS EARNING
A one-minute explainer of the idea behind this story: useful life. It teaches the concept, not this story’s figures.
The key insight: NVIDIA’s blog names useful life as one of three things that shape an AI factory’s return, and the sources it cites measure it in different ways: Sprout reports Microsoft running several GPU generations for 6.9 to 8.8 years, and Ornn reports the five-year A100 rental price at 80.2% of the one-month price against 44 to 60 percent for Hopper and Blackwell families. NVIDIA’s post also cites Silicon Data, which this publication could not open, for a six-year-old A100 “still worth a quarter of what it cost”.
NVIDIA’s post of 1 October 2026, “Productive, Durable, Fungible: How NVIDIA AI Factories Maximize Return on Investment,” says “Each megawatt factory costs roughly $60 million.” It says three things shape a factory’s return: earning capacity, useful life and demand.
NVIDIA says its platform is productive, durable and fungible, which it says maximizes all three. To support that, the post cites SemiAnalysis, Sprout, Barkr, Silicon Data and Ornn Data. This publication opened three of the five originals and could not open the other two.
The $60 million per megawatt figure is NVIDIA’s, and this publication did not check it.
Ornn’s five-year term price as a percentage of the one-month term price, by chip family. Ornn describes the marks as analyst-produced indicators, not executable quotes.
The Throughput Claim and the Engine Label
The post says SemiAnalysis AgentX data shows Vera Rubin NVL72 systems “deliver over 30x higher throughput per megawatt than NVIDIA GB300 NVL72, and up to 45x lower cost per million tokens on the DeepSeek V4 Pro model.”
NVIDIA’s earlier post on this, dated 24 August and updated on 15 September 2026, worded the throughput line as “up to 30x higher throughput per megawatt.” The 1 October text says “over.” Neither version gives an interactivity target or a serving engine.
SemiAnalysis’s written post of 14 September does give them, and the multiple depends on both. For DeepSeek V4 Pro it reports Rubin at about 2.09x the stronger GB300 engine at a 100 tokens-per-second interactivity target, about 7.2x GB300 SGLang at 150, and 2.72x at 200.
At exactly 170, SemiAnalysis reports 62.9x the throughput per megawatt of GB300 TRTLLM and 5.56x against GB300 SGLang. It adds: “The engine label is therefore essential when quoting the high-interactivity gain.”
SemiAnalysis also says the size of the advantage “depends on the interactivity target and the serving engine used for comparison.” This publication did not check the 45x cost figure against SemiAnalysis, whose result tables appear as images.
How Long a GPU Earns: What Sprout Says
NVIDIA says every major operator has extended server life and cites a Sprout chart. Sprout’s own post of 30 September says Microsoft ran several GPU generations for 6.9 to 8.8 years, and that its V100 fleet ran 8.4 years against a six-year book life.
Sprout also says AWS has never retired an A100 and that rental rates for prior-generation GPUs rose through 2026. It says six-year-old A100 systems are selling at roughly 38% to 43% of their launch price, above the 25% residual now used in lending.
NVIDIA’s post gives a different measure from Silicon Data: a six-year-old A100 “still worth a quarter of what it cost.” This publication could not open the Silicon Data post, so it cannot compare the two.
The 80 Percent: What Ornn Says
NVIDIA says Ornn Data finds “the market paying 80% as much to rent an A100 GPU on a five-year contract as on a one-month contract.” Ornn’s paper, “The Economics of Open-Weight Inference,” reports the A100 figure as 80.2%, from marks published on 13 August.
The same paper gives the other families. At five years the H100 is 59.8%, the H200 43.7%, the B200 53.8% and the B300 53.8%, which Ornn summarises as 44 to 60 percent for Hopper and Blackwell. NVIDIA’s post quotes the A100 figure only.
Ornn describes the marks as “analyst-produced indicators, not executable quotes or verified averages of comparable executed contracts at every tenor.” It also says percentage retention depends on each family’s starting level, so a flatter curve “need not mean a higher absolute rent.”
Ornn’s disclosures say it licenses data commercially and also offers GPU rentals through Ornn Compute, which it says creates “a commercial interest in both the interpretation and adoption of its data.” Its own diagram lists continued earning capacity of older accelerator families as an “economic inference to be tested.”
NVIDIA says Barkr puts useful life at five to six years for an eight-GPU H100 system and nine to 10 years for GB300 NVL72, based on resale. Barkr’s report page lists a 16-page PDF dated 29 September, “The Resale Standard: Forward-Looking GPU Valuations,” which it describes as forward valuation curves for the H100 and GB200 NVL72.
This publication did not open that PDF, so it cannot say where the GB300 figure comes from. The Silicon Data claim above is also unchecked.
The Structural Read
The sources measure different things. Sprout reports how long operators ran hardware and what old systems sell for; Ornn reports rental prices at different contract lengths as a share of the one-month price; the Silicon Data figure was not opened. The piece sets them side by side without converting one into another.
Ornn’s own caveats are in the piece: it describes the marks as analyst-produced indicators, not executable quotes, says percentage retention depends on each family’s starting level, and discloses a commercial interest through Ornn Compute. NVIDIA’s post quotes the A100 figure only.
On the throughput line, the originals add what NVIDIA’s sentence omits: SemiAnalysis gives an interactivity target and a serving engine, and says the engine label is essential when quoting the high-interactivity gain. NVIDIA sells the hardware it discusses, and this piece did not check the 45x cost figure.
NVIDIA blog, 1 October 2026
“Three key things shape AI factory returns: 1. Earning capacity: What the factory could earn in a year if it sold every token it can produce. 2. Useful life: How long its AI hardware keeps earning. 3. Demand: How much demand there is for those tokens.”
Three Implications
ONE TERM, THREE YARDSTICKS. Useful life appears in NVIDIA’s post as one of three drivers of return, and the cited sources measure it by years in service, resale price or rental retention. The piece does not say which yardstick is right.
THE SELLER FRAMES THE QUESTION. NVIDIA sells the hardware it discusses, and the piece reports where the originals add detail or caveats the post leaves out, such as the engine label and Ornn’s disclosures.
WHAT REMAINS OPEN. The $60 million per megawatt figure, the CoreWeave booking claim, the Silicon Data and Barkr figures and the 45x cost figure are unchecked, and the piece does not judge whether an AI factory earns a good return.
What Is Not Established
The $60 million per megawatt figure, the claim that CoreWeave extended bookings on 2020 units through 2029, the Silicon Data figure and the Barkr figures are NVIDIA’s statements and are unchecked here. The 45x cost figure is unchecked too.
Nothing here judges whether an AI factory earns a good return. NVIDIA’s post is a company blog that sells the hardware it discusses, and every figure above is attributed to its source. Nothing in this piece is investment advice.
A note on sourcing. We read NVIDIA’s blog posts, the SemiAnalysis post, Sprout’s post, and Ornn Data’s paper and Barkr’s report listing page against NVIDIA’s claims. We did not open Silicon Data’s post or Barkr’s PDF, so those two sources remain unchecked. Every figure here belongs to the party that published it, and we have not checked any of it independently. NVIDIA sells the hardware it discusses. Nothing we have written is a forecast or investment advice.
Gennaro Cuofano is a CRO and tech executive who has worked in AI since late 2015, starting with bringing NLP, natural-language generation, voice and chatbot products to the marketing industry. His background is in law and finance: he holds a Master's degree in Law and an International MBA with an emphasis on corporate finance (LUISS Business School and the University of San Diego, 2012), and worked as a financial analyst at a real-estate investment firm in San Diego and as an assistant controller. His work focuses on business model strategy, business engineering and, more broadly, structural analysis: how companies make money, read from their own filings. He created FourWeekMBA and leads research there and at The Business Engineer, his newsletter on AI and business strategy, with over 95,000 subscribers and more than 1,000 published analyses. His writing has also appeared on Entrepreneur (for example, How to Digitize a Family Business in Europe) and HackerNoon. Find him on LinkedIn and Substack. See how we source.
