A read-only audit (fully creditable) and a fixed-price pilot against your live operation. You commit to the multi-year partnership only on proven, measured results.
The platform is stood up against your systems and proven on real workflows. Your data and fine-tuned weights are yours from day one.
You run on an annual licence while we operate and continuously improve the platform, and your engineers ride along, so knowledge transfers with the code.
You exercise a priced buyout and take ownership: a clean, unencumbered fork of the platform, the weights and the infrastructure. Licence fees to us end.
The licence is evergreen, it does not expire into a cheap hand-over, because that would make waiting out the term the rational move and collapse the structure. And the transfer is a priced buyout, set as a multiple of the annual value the platform is independently measured to create for you. You capture roughly three-quarters of the value the platform generates; the buyout is a fraction of what you would pay to rent equivalent intelligence forever.
You save $250.7M over ten years, and own the platform from year 4, a capitalised asset instead of perpetual rent.
Illustrative. Buyout priced as a multiple of independently verified annual value; maintenance ~18%/yr post-transfer. Your Recon produces the real figures.
It is a three-stage commercial model. In Build we stand the platform up and prove it on a real workflow; in Operate you run on an evergreen annual licence while your engineers ride along, so knowledge transfers with the code; in Transfer you exercise a priced buyout and take ownership. Read the full argument at build, operate, and own.
The transfer is a priced buyout, set as a multiple of the annual value the platform is independently measured to create for you, not a number pulled from the air. You capture roughly three-quarters of the value the platform generates, and the buyout is a fraction of what renting equivalent intelligence forever would cost. Because the value is measured, the price is arithmetic rather than negotiation theatre.
Because a licence that expired into a cheap hand-over would make simply waiting out the term the rational move, collapsing the whole structure. An evergreen licence keeps the transfer a deliberate, priced decision, so ownership is something you choose to buy, not something you run out the clock to obtain.
You own it. At transfer you take a clean, unencumbered fork of the platform, the fine-tuned weights and the infrastructure, and licence fees to us end. That makes it a capitalised asset on your balance sheet, not perpetual operating expense. See how ownership works.
Yes. The weights are fine-tuned on your data, inside your perimeter, throughout the engagement, and they transfer to you with the code, they were never ours to withhold. You leave with the model your own operation has been sharpening month after month, not a generic base you have to start improving from scratch.
Far less than the frontier-lab path, which asks for tens of millions in year one before working software exists. You start with a fully creditable read-only audit and a fixed-price pilot scoped to a single workflow, and commit to the multi-year partnership only on proven, measured results. Indicative figures are quoted against your Recon findings; payback is designed to arrive inside twelve months.
The whole model is built to remove that risk. Because the platform runs inside your perimeter and the engagement ends with you owning a clean fork of the code and the weights, you are never dependent on our continued existence to keep operating. Source-code escrow and continuity terms are written into the agreement so that even mid-engagement you retain a path to the running system.
Renting from a frontier lab means your data on their servers and a bill that never ends; building in-house means years of platform engineering before you touch a workflow. Limen ships a governed, sovereign system in weeks, improves it inside your perimeter, and hands you the asset at the end. You get the speed of buying and the independence of building. See the economics.