CME Group and Silicon Data are set to launch two futures contracts today tied to the hourly rental cost of AI graphics processors — the first regulated derivatives designed to let companies hedge the price of the computing power behind artificial intelligence.
The contracts, the Silicon Data H100 Rental Index Futures and the Silicon Data B200 Rental Index Futures, will track indexes measuring hourly rental GPU costs published by Silicon Data, according to the companies' announcement. Each contract will represent a month's worth of rent for Nvidia's H100 chip and the next-generation Nvidia Blackwell B200 respectively, and the contracts will be listed under the rules of NYMEX.
Silicon Data is described by the companies as a leader in GPU market intelligence and benchmarking, and is backed by the global trading firm DRW. The launch, announced on August 11, was planned for October 5, 2026, pending regulatory review.
Why it matters
Explosive demand for AI infrastructure has driven sharp swings in compute prices, leaving AI developers and hyperscalers exposed with no standard way to lock in costs. "For years, two companies buying the exact same GPU capacity could pay wildly different prices with no way to know who got the better deal," said Carmen Li, chief executive officer of Silicon Data. "Compute futures give the market something it's never had: a public, tradable reference price for the resource every AI system runs on."
Pete Keavey, CME Group's global head of energy and environmental products, framed the move in historical terms: "Compute has become the currency of the AI age... Just as oil fueled the 20th century economy and evolved from spot trading into a global derivatives market, our futures contracts will now turn compute into a standardized, tradable commodity that will provide global businesses with a reliable, regulated venue to manage price risk."
The contracts are also expected to give investors a window into expected future AI spending, turning the opaque market for AI compute into something markets can price, plan around, and bet on.
Sources:
CME Group and Silicon Data to Launch Compute Futures on October 5 — PRNewswire via BitcoinWorld
The contracts, the Silicon Data H100 Rental Index Futures and the Silicon Data B200 Rental Index Futures, will track indexes measuring hourly rental GPU costs published by Silicon Data, according to the companies' announcement. Each contract will represent a month's worth of rent for Nvidia's H100 chip and the next-generation Nvidia Blackwell B200 respectively, and the contracts will be listed under the rules of NYMEX.
Silicon Data is described by the companies as a leader in GPU market intelligence and benchmarking, and is backed by the global trading firm DRW. The launch, announced on August 11, was planned for October 5, 2026, pending regulatory review.
Why it matters
Explosive demand for AI infrastructure has driven sharp swings in compute prices, leaving AI developers and hyperscalers exposed with no standard way to lock in costs. "For years, two companies buying the exact same GPU capacity could pay wildly different prices with no way to know who got the better deal," said Carmen Li, chief executive officer of Silicon Data. "Compute futures give the market something it's never had: a public, tradable reference price for the resource every AI system runs on."
Pete Keavey, CME Group's global head of energy and environmental products, framed the move in historical terms: "Compute has become the currency of the AI age... Just as oil fueled the 20th century economy and evolved from spot trading into a global derivatives market, our futures contracts will now turn compute into a standardized, tradable commodity that will provide global businesses with a reliable, regulated venue to manage price risk."
The contracts are also expected to give investors a window into expected future AI spending, turning the opaque market for AI compute into something markets can price, plan around, and bet on.
Sources:
CME Group and Silicon Data to Launch Compute Futures on October 5 — PRNewswire via BitcoinWorld
