355
Audio & Video Production344
Automation & Workflow224
Software Development250
Marketing & Growth192
AI Infrastructure & MLOps173
Writing & Content Creation203
Data & Analytics140
Design & Creative169
Customer Support130
Photography & Imaging156
Sales & Outreach125
Voice & Speech135
Operations & Admin87
Education & Learning131
CME Group says it will launch futures contracts based on GPU rental rates, giving companies and traders a way to lock in future AI computing costs.
In short: CME Group plans to launch futures contracts that track the future rental price of GPUs, a key type of computing power used to build AI.
CME Group, one of the world’s largest markets for financial contracts, says it plans to launch a futures market linked to AI computing power later this year. The contracts are designed around the future rental cost of graphics processing units, or GPUs, which are specialized chips often used to train AI systems.
GPU supply can be tight and orders can take months. Rental prices can also move up and down quickly. CME says the new contracts would let traders and companies either bet on where rental prices are going or protect themselves from price jumps.
CME is partnering with Silicon Data, a company backed by trading firm DRW. Silicon Data provides pricing indices, meaning published reference prices that work like a scoreboard for typical GPU rental rates. The futures contracts will be based on those indices, including daily rental rates for Nvidia chips such as A100, H100, and B200 that appear on platforms like Bloomberg and LSEG.
Many AI products depend on renting large amounts of computing power. A futures contract is a way to agree on a price now for something you will buy or sell later, like locking in the future price of fuel for a delivery fleet. If these GPU futures take off, they could make it easier for AI companies, banks, and investors to plan budgets and manage uncertainty when computing costs swing.
Source: Financial Times