Home Platform Solutions Operators Investors Resources Blog Request Briefing
Solutions

Two verified products. One settlement rail.

RAW settles delivery across two asset classes — AI inference and GPU compute — both attested by the same on-site device and settled on the same Canton Network rail.

Settlement units

Built for asset classes that already generate continuous, measurable output

In short

RAW settles two kinds of compute delivery — AI inference and GPU compute — using one attestation device and one settlement rail. Each product answers a different "did this actually happen" question for its buyers; both settle atomically in regulated stablecoins once verified.

Primary Product

Verified AI Inference

Measured in millions of tokens per hour — input and output — with throughput and TTFT attested at the edge.

Settlement unit1M tokens
Quoted asMTph (M tokens/hr)
VerificationRAW Box edge-attested
Base Physical Unit

Verified GPU Compute

The attested GPU-hour: one hour of a specified accelerator, device class, and configuration, delivered under contract by a named, attested asset. Heterogeneity is disclosed and priced at contract formation — not discovered after delivery.

Settlement unitAttested GPU-hour
Specified atDevice class + config
VerificationRAW Box signed telemetry
The two settlement products, side by side
ProductSettlement unitVerified atVerification method
Verified AI Inference 1M tokens (MTph) Edge (RAW Box) Signed telemetry; throughput and TTFT attested at inference
Verified GPU Compute Attested GPU-hour Edge (RAW Box) Signed telemetry; device class and configuration disclosed at contract formation
Market context

The world's largest derivatives exchanges launched compute futures. Verified physical delivery remains unclaimed.

Both futures markets are cash-settled against price indices. Neither specifies a delivery-verification mechanism. RAW is the physical leg.

~$106B
AI inference market, current
~$255B
Projected 2030
~$600–725B
Hyperscaler capex 2026
~5%
Avg enterprise GPU utilization

Inference market figures: Grand View Research, 2025–2030 forecast.

By use case

Who settles on RAW

Inference Providers

Monetize idle GPU capacity

Serve verified inference under your own brand while every delivery claim is independently attested at the edge before payment moves.

Data Centers

Turn utilization into a financial asset

A verified track record of delivered GPU-hours becomes portable, financeable proof — visible to lenders and institutional buyers, not trapped in private billing systems.

Institutional Buyers

Buy output, not a price proxy

Get programmable, verified exposure to compute output instead of a cash-settled price index — on the same regulated rail your counterparties already use.

FAQ

Frequently asked questions

It follows what your facility already produces: AI inference if you serve model traffic, GPU compute if you rent out accelerator time in bulk. Most facilities settle under one product to start.
Yes. A data center that both serves inference and rents raw GPU-hours can settle both under the same RAW Box and the same Canton Network contract framework, since attestation and settlement are shared infrastructure across both products.
Price is fixed at contract formation, not discovered after delivery. For GPU compute specifically, device class and configuration are disclosed and priced up front, so heterogeneity is a contract term rather than a post-delivery dispute.
No. Both settle through the same on-site attestation device and the same Canton Network settlement rail. The difference between products is what's measured and how it's verified, not how the buyer connects.
Talk to the team directly. The product set reflects the two asset classes with the clearest, most immediate verification path today; facilities with mixed or unusual output are exactly the case a briefing is meant to work through.
RAW
Get access

Find the right settlement product for your asset