Injective launches perpetual futures tracking five Nvidia GPU rental costs

This new index enables leveraged trading and provides a mechanism for hedging against shifts in the costs associated with AI computing.

The SICI market is now live on the mainnet, utilizing USDC as its margin, and is designed to integrate into various applications across the Injective ecosystem, according to the October 26 announcement. The benchmark itself is constructed and priced by LO:TECH, a capital markets firm based in London, utilizing rental listings sourced from compute marketplaces.

LO:TECH characterizes this product as a means for participants to take positions on computing expenses, while also allowing AI developers and infrastructure providers to hedge their financial exposure. The firm noted that escalating rental bills represent a risk for developers, whereas declining rental income poses a risk for equipment providers.

The benchmark weighting includes 40% allocated to Nvidia’s H100 SXM 80GB, followed by 25% for the B200, and 15% for the H200. The remaining 10% is split equally between the A100 SXM 80GB and the GeForce RTX 4090.

The inclusion of the consumer-grade RTX 4090 reflects the reality that developers rent these cards for smaller AI workloads and running open-source models. The methodology dictates that each component maintains its assigned weight within a specific version, ensuring historical index calculations remain preserved should the methodology change.

This five-chip product follows Injective’s prior H100 rental-price contract, which was launched in August 2025 in collaboration with Squaretower. For the SICI, LO:TECH aggregates reference prices in dollars per GPU-hour from compute marketplaces, combining both on-demand and spot listings.

The calculation process involves rebasing each GPU’s reference price to 100 based on the July 22, 2026 daily fixing. LO:TECH then combines the five normalized component levels using the established weights before publishing a trailing seven-day average of the hourly basket. The SICI is quoted in index points, though its underlying inputs are hourly rental prices.

This development occurs against a backdrop where institutional participants are actively seeking regulated exposure to computing costs. For instance, the Hyperliquid Policy Center submitted an October 20 request to the Commodity Futures Trading Commission seeking regulated contracts tied to GPU rentals and AI model operational expenses.

These market developments reflect a broader trend where financial institutions and decentralized platforms are creating structured financial instruments to address the volatility inherent in the rapidly expanding AI infrastructure sector.

Why it matters

This launch brings regulated, onchain exposure to the rapidly fluctuating costs of AI computing hardware. It mirrors broader industry interest, such as Hyperliquid Policy Center's October 20 submission to the CFTC regarding compute perpetuals. These developments signal increasing institutional focus on monetizing and hedging AI infrastructure risk.