Using transaction data from long-term contracts, @OrnnExchange has modeled a price surface for GPU rentals as a function of order size and contract length. The data shows a positive correlation between price and GPU capacity size, and an inverse correlation between price and contract length. Now, what's the underlying reasons behind these trends? 1/ Small orders are filled from leftover capacity. As a result, sellers may discount it to maximize utilization. 2/ Big buyers are buying scarcity. A request for >1k H100s, etc clustered together has only a few vendors to choose from. See the numbers at and rent your capacity at

The case for compute markets and compute indices begins with a growing imbalance: the ambition to scale AI depends on enormous commitments of capital, while the people financing that expansion face extreme market uncertainty. Building lasting AI infrastructure requires a better way to understand, price, and share that risk. Compute has become a strategic resource. Access to it shapes who can develop advanced AI, bring new products to market, and compete on a global stage. For companies and nations alike, securing reliable compute capacity is increasingly tied to economic opportunity and technological independence. The opportunity we are attacking is building the market infrastructure that allows more people to participate in AI’s expansion with confidence. That means giving a developer greater visibility into future compute costs, an operator a stronger basis for planning capacity, and a financier a clearer understanding of the exposure they are taking on. We have an opportunity to make the economics of compute more transparent and its risks more manageable. Doing so will support the next generation of AI infrastructure.

An NVIDIA GPU is a high-yield asset. An H100, bought on the secondary market in Sep 2025 for ~$19k: Resale today: ~$20.5k Net rent earned: ~$13.5k Total: ~$34,000, +76% in 1 year. A B300, bought in Jul 2026 for ~$54k: Resale today: ~$51k Net rent earned: ~$14k Total: ~$65,500, +21% in under 3 months What allows these assets continue to earn? Resale value has held as rental rates have increased and demand has grown across all hardware generations. Resale rates have held within −5% to +7%, while Ornn's H100 index is $2.91/hr, +49% from a year ago, and our new B300 index tracks at $11.32/hr, up 66% from June. H100 marketplace utilization is 87% today. B300 is at 90%. Ornn's forward curves predicts that in 2 years, by Sep 2028: H100 ~$63,500 (+227%) B300 ~$186,000 (+244%) NVIDIA Hopper and Blackwell hardware both hold incredible residual value.

Can the rental price of GPUs predict stock prices? Ornn's A100 GPU rental index has led Astera Labs's stock price by about 5 weeks. - The index's 20-day moves line up with ALAB's 20-day moves 25 trading days later (r = 0.74). - Moved forward 5 weeks, the index tracks ALAB's price at r = 0.87 after removing the shared upward trend. - It holds across the year, and it holds even after stripping out the chip sector's own moves or NVDA's. Astera Labs makes the connectivity chips inside AI servers. GPU rental prices show how tight AI capacity is right now. The companies that sell into AI server build-outs are priced on that tightness after it shows up in orders. Ornn prices compute where it actually trades. The signal shows up here before it shows up anywhere else.

The open vs closed source race is a price war. Since June, the average paid token from OpenAI got 62% cheaper. DeepSeek: −51%. Google: −29%. Anthropic: −11%. Over the same three months, Ornn has tracked B200 rent rising 50%. H200 rose 28%. Models are getting cheaper to buy. The compute to run them is getting more expensive to rent. So where does value accrue? To whoever owns the compute. The cheapest open-weight model finishes a task for about 1/5 the cost of a comparable closed model, and they don't need the newest chips. Every GPU is a shovel when the models are free. See more at

$NBIS's new rates sit 4 to 49 percent above what those chips rent for on OCPI, H200 closest at $5.40 against $5.19 and H100 widest at $4.50 against $3.03. B300 has no rental price on the index, and the forward curve prices it at $7.45 one month out against a $9.50 list.

Live rental prices, by accelerator: B200 $7.09 H200 $5.40 H100 $2.72 RTX PRO 6000 $1.50 A100 $1.05 RTX 5090 $0.68 Each GPU is a separate market with its own supply and demand. An hour of B200 costs 10.4 times an hour of RTX 5090. See more:

B200 PRICE UPDATE (09/15/26): A B200 rents for $7.24 an hour, up 21.7 percent from $5.95 on August 15. Of the B200s we track, 11.6 percent sat unrented today. Over the same month the H100 fell 10.9 percent, from $2.84 to $2.53.

Today, we are introducing the Ornn B300 Payback Forecast. A GPU pays for itself through rental income. Until now, nobody could tell you how long that takes for a chip that only started shipping this year. Built on OCPI forward curves, this forecast projects how long a B300 takes to earn back its purchase price at market rental rates, giving operators, lenders, and investors a repayment timeline they can underwrite against. It's live now on Ornn Data, see it here:

After such a consequential week for AI, with the leading labs demonstrating real progress, today's news feels especially timely: @SVAngel is launching Project Blueprint, a comprehensive effort to forge consensus among industry leaders and lawmakers around durable solutions to the hardest AI policy challenges so that all Americans can benefit from AI's opportunities. And we've brought on @JayCarney to run it.  We are encouraged by the reaction of Sam Altman, Dario Amodei and Demis Hassabis to the launch of Project Blueprint: Sam Altman, co-founder and CEO of OpenAI: "We have before us a technology of immense and still-unfolding wonder. I believe AI has the power to heal people, to discover cures and to deliver abundance on a scale the world has never known before. But we need a way for the world to build trust in the technology so that we can all get to share these benefits. National safety requirements for the most capable systems are a great first step, ideally building towards a global framework for advanced AI models. I’m glad Ron is bringing the people building AI and policymakers together to help move that work forward.” Dario Amodei, CEO and co-founder, Anthropic: "AI is advancing faster than our institutions can adapt, and the most important decisions about how it is developed and deployed will be made in the next few years. If those decisions are to serve the public, they can't rest with AI companies alone. Industry and policymakers need to work through them together, with an honest, shared understanding of what these systems can do. We're glad to support Project Blueprint's effort to make that possible."  Demis Hassabis, chairman and co-founder of Google DeepMind and the chief scientist of Alphabet: "We have long said that AI development needs to be both bold and responsible. We support bringing industry and lawmakers together around public policies that promote both innovation and safety."

Today, we are introducing the Ornn B300 Payback Forecast. A GPU pays for itself through rental income. Until now, nobody could tell you how long that takes for a chip that only started shipping this year. Built on OCPI forward curves, this forecast projects how long a B300 takes to earn back its purchase price at market rental rates, giving operators, lenders, and investors a repayment timeline they can underwrite against. It's live now on Ornn Data, see it here:

OpenAI's Navier-Stokes run used about 130 billion output tokens. At the flagship list rate of $50 per million that reads as a $6.5 million bill. Against Ornn's B200 settle of $6.84 an hour, that same $6.5 million is 950,000 GPU-hours, a thousand-GPU cluster running for forty days.