Pricing
We price Worldmodel the way enterprise software should be priced: by organizational scope, not by seats. Seats are the wrong mental model for a model of your company. The value of a world model scales with the complexity of the organization, not with the number of humans who log in.
Structure
Foundation. A one-time scoping and deployment fee covering the initial build of your world model: security review, data source integration, entity resolution, and go-live. This is not a setup fee hidden in the subscription. It is a real engagement, typically four to eight weeks, that produces a functioning world model tailored to your organization. Sized to scope.
Platform. An annual subscription for the Worldmodel platform itself. Priced by organizational scope: single business unit, multi-business unit, or full enterprise. The subscription includes the world model, continuous reasoning, base compute envelope, and all platform updates. The price reflects the complexity of what the model needs to understand, not how many people use it.
Workers. Each AI worker is priced separately. Every worker comes with a bundled compute envelope sized to its role. Overages are billed transparently at cost-plus, with full visibility into usage. No hidden margins on compute.
Bring your own models. Many organizations already have contracts with model providers. Worldmodel supports connecting your existing API keys from Anthropic, OpenAI, Google, or any other provider. When you bring your own models, you pay us only for the platform and the worker logic, not for inference. Your tokens run on your accounts, under your rate limits, governed by your existing agreements. We include a bundled compute envelope for organizations that prefer a single vendor relationship, but we never require it.
Philosophy
We don't publish prices because enterprise pricing should be a conversation, not a web form. The scoping, deployment, and organizational complexity of every customer is different, and pretending otherwise would mean either overcharging simple deployments or underpricing complex ones.
Contracts are always annual. At this tier, monthly billing creates the wrong incentives for both sides. We want customers who are committed to building a world model, not experimenting with a subscription they might cancel.
Compute costs are transparent. Organizations that bring their own model provider accounts run inference on their existing contracts at their negotiated rates. For those who use our bundled compute, we publish the exact margin we charge on top of provider costs. Either way, there are no hidden fees on inference.
We do not offer free pilots. We offer a paid eight-week Discovery engagement that produces a deployable world model scope and a working worker prototype. Discovery is real work that produces real output. It is designed so that at the end of eight weeks, you have enough to make a deployment decision, and we have enough to give you an honest price.
For pricing specifics, contact us at contact@theworldmodelcompany.com.