Pricing Plans

Start with Free to try every AI model. Starter removes the watermark and adds Batch Run. Pro adds parallel GPU processing and larger batches for heavy workloads. Business gives your team a shared workspace, and Enterprise adds your own deployment, SSO and dedicated support.

Yes. Upgrade or downgrade from this page whenever you like, and use Manage Subscription to update your payment details or cancel. Monthly plans have no long-term commitment.

With yearly billing you pay for twelve months up front, at a lower price than paying monthly. The saving is shown next to the Monthly / Yearly switch, and the cards show the monthly equivalent.

Yes. Eligible workspaces can try Pro free for two weeks, with no credit card required. If you add a payment method, your subscription continues after the trial; if you don’t, the trial simply ends and you stay on Free. Each workspace can use the trial once.

Payments are processed securely by Stripe. CNAPS never sees or stores your full card details.

Open Manage Subscription on this page to see your billing history, download invoices and receipts, and update your billing details.

GPU time is the processing time AI models spend on your inputs. Every plan includes an allowance of GPU time that refills automatically every few hours, so each window starts fresh.

The model and your inputs. Higher-resolution images, longer videos and heavier settings take longer to process. For example, upscaling an image 4× uses far more GPU time than classifying it.

No. When a window ends, unused GPU time is cleared and your full allowance is restored, so every window starts from the same amount.

You can start new runs again after the next refill, or upgrade for a larger allowance right away.

Queued runs are how many runs you can start at once: they wait their turn instead of being turned away. Parallel GPU processing, from Pro up, runs more of them on GPUs at the same time, so results come back sooner.

Batch Run sends many inputs through the same flow in one go, such as a whole set of product photos. It is available from Starter, and Pro and above handle larger batches.

All of them, on every plan — Free included. Plans differ in how much GPU time you get and how fast runs are processed, not in which models you can use.

No. You can create and save as many flows as you like on any plan.

Yes. Outputs created on Free carry a watermark. Every paid plan removes it.

Yes. On any plan you can publish a flow to the Community, where others can discover it and fork it into their own workspace.

Yes, on every plan. Create an API key to use CNAPS from automation tools such as n8n or Zapier, or connect AI assistants such as Claude through MCP to build and run flows from a chat.

MCP (Model Context Protocol) is an open standard that lets AI assistants use outside tools. Connect CNAPS to an assistant such as Claude, and you can ask it in plain language to find models, build a flow and run it for you.

Business runs in a separate team workspace. When you choose Business, you create a team workspace or pick an existing one, and your personal workspace stays as it is.

Business includes a set number of seats, shown on the plan card, and each additional seat is billed separately. Everyone in the workspace shares the team’s GPU time and flows, with roles that control who can manage members and billing.

Three: dedicated SaaS, an enterprise environment on our service; private cloud, a cloud deployment configured just for your company; and on-premises, fully air-gapped with no outbound connections. Contact us to find the right fit.

Yes. Enterprise plans can include single sign-on with SAML or OIDC, SCIM user provisioning, and help with your security and compliance review, set up as part of your contract.