Budget each rollout phase. Know who pays for what.
One fee for the delivery scope; running costs in your budget. Nxgyle invoices the design and delivery work agreed for each phase. After handover, workers carry no run-time fee and no usage charges from us. Hosting, model usage and maintenance sit with the teams that run them.
Fix scope, owners and handover terms up front.
Your program budget covers an agreed scope: the processes and connected systems Nxgyle designs and delivers. Volume in operation never changes that fee. Managed Delivery can be added for support, but it is never required. Changes start with a request from your process owner, and after handover your own team maintains each worker — and keeps the code.
A fixed scope for each phase
Each phase takes named workers from design through trial runs to delivery. The fee follows which processes and systems that phase covers, and both are agreed before work starts. The number of cases the workers later handle never changes it.
Support only if you add it
Managed Delivery is a support service your program can add; no worker depends on it. Nxgyle can handle upgrades and changes while your operators run the workers, or your engineers take both over at handover. Every export includes its tests, and runbooks or monitoring can be added.
Changes your owners request
When a process or policy changes, the process owner asks for a new version; no subscription sets the schedule. An optional upgrade feature carries your engineers' code edits into the regenerated worker and lists any conflicts for review.
Each phase is priced from its scope, so no list price appears here.
A published rate cannot reflect your volumes, integrations or control requirements, and the site-wide claims policy applies to pricing as well. Your figure comes from a scoping workshop with your process owners and program lead, not from a rate card.
White-label business applications, generated and delivered complete.
Each runs one defined process under fixed rules, and its source code is yours.
Catalog coverage spans finance, healthcare, manufacturing, energy, legal, logistics and automotive.
Each count is calculated from the portal and worker records listed on this site. A digital worker is an operational asset in code: it runs one defined process under your rules and approval checkpoints, and your organization holds the full source.
Nothing to renew, meter or report.
Your engineers can confirm each one in the delivered code, on your own systems, before the program signs off.
No usage charges
Nxgyle bills nothing for the cases an exported worker handles. Your model vendor invoices its calls directly, and hosting and maintenance stay with the teams that run them. Budget those costs at the volumes each phase expects.
What each export lets you checkNo license check
Exported workers run from source code on systems you host, and there is no license key or activation step to administer. This applies to exported workers only; Nx·Studio and Nx IDE carry their own licenses.
What each export lets you checkNo calls to us
A worker in operation never contacts Nxgyle, so no usage is counted and nothing can be switched off from outside. The signed manifest in each export lists its AI models, agents, connectors and code libraries, and your engineers verify it on your systems.
What each export lets you checkYour engineers check the signed bill of materials in each export offline, with no call to Nxgyle. Before the program commits, your risk lead can ask the Trust Center for a redacted example.
How design choices set the run budget.
Wherever possible, an explicit business rule becomes a fixed rule check instead of an AI call, and that step carries no model cost at all. Nothing is discounted; the step simply never calls a model. Before the build, each design option shows estimated hours saved and running cost per month; the generation plan then adds a payback estimate your finance partner can compare.
Steps that still need AI judgment use the model accounts your organization contracts, local models on your own hardware included. A step cap and a time limit can be set for each run. Your model vendor bills that usage at your contracted rates, with nothing added by Nxgyle. For routine decisions that repeat all day, our Cognitive Engine can settle qualified cases without a model call. It runs outside Nx·Studio workers, and its Studio provides the decision API.
No savings percentage appears here. The mechanism is published instead, so each phase's budget rests on your own volumes.
Treat the delivery fee and the run budget as two lines with two owners. Plan model usage, hosting, exception review and maintenance together. Any optional support is agreed around your service needs.
- Nxgyle program fee Invoiced per phase against the agreed processes and connected systems, fixed before work begins.per phase
- Steps run by rules Rule steps make no model call, so they carry no AI cost. Anything Nx·Studio cannot convert is flagged for your process owner.zero
- AI judgment calls Billed by your model vendor at contracted rates, with nothing added by Nxgyle. Every design's run cost is estimated ahead of the build.vendor
- Run-time environment Hosting your IT team runs and budgets for. No license check applies, and nothing is sent to Nxgyle.IT budget
- Managed Delivery Charged by Nxgyle only if your program adds it.if added
No figures appear here: scope, not usage, sets the fee, and any generic figure would misstate your program budget.
If the engagement ends, what keeps running?
Your exported workers continue to run, because none depends on Nxgyle: there is no runtime, activation check or usage meter. Your team keeps the source code and the signed bill of materials. From then on, operation depends on your infrastructure, third-party services and maintenance.
Contents of each export packageWhich work does the Nxgyle fee pay for?
The fee covers a scoped engagement in which your digital workers are designed, trialed and delivered. Nxgyle adds no usage charges and no run-time fee. Explicit business rules turn into fixed checks with no model call, and the remaining judgment runs on model accounts your organization contracts. Before anything is built, you can see what each design is estimated to cost to run.
What makes up run costsWhy don't you publish a rate card?
Each phase is priced from the processes and systems it covers, so publishing a list price would misstate your case. The site-wide claims rule applies here too: a number that cannot be stated honestly is not published. Your figure comes out of a scoping workshop with your process owners.
Book a scoping workshopIf case volumes double, does our Nxgyle fee change?
No. The fee stays where the engagement set it, because volume creates no usage charge from Nxgyle. Model-vendor and hosting costs do rise with volume, and extra delivery or support scope is agreed separately. Size each phase's run budget from the design estimates and the measured cost of trial runs.
Must the program include Managed Delivery?
No. Your engineering team can maintain exported workers, or you can scope Managed Delivery for ongoing support. Every export includes its tests, and runbooks or monitoring can be added. Agree release ownership, support duties and service scope before handover.
What does our team take over at handover?
Each worker's code, blueprints and test suite, plus any dashboards or deployment files you opt into. Wherever approval checkpoints exist, a reviewer inbox app comes too; your engineers wire it up. It all runs in your operating environment on the cloud and model vendors you select, with no license keys and nothing at run time that depends on Nxgyle.
Review our controls