Build complete AI workflows free on Creator. Move to a private deployment when you need dedicated compute, your own packages, custom connectors, or the workflow running inside your own infrastructure.
For individual builders, prototypes and experimentation
Runs on shared infrastructure, with a 5 MB file upload limit and the built-in Python libraries.
For teams that need private infrastructure, dedicated compute or production deployment
Sized to your workload — not a seat count you have to grow into.
For consultants, MSPs and agencies building workflows for several clients
Applications are open — we are taking five founding partners to begin with.
Creator is free forever — not a trial, no expiry, no card. A private deployment is quoted against your deployment, your integrations and the compute you actually need. There are no artificial tiers in between.
Start on Creator. It costs nothing, and the limits below are the honest boundary — when you hit one, that is the moment to talk to us.
Ask us a question →Implementation, integration and training are delivered by Zapwiser, the consulting company behind QMerse — quoted as a project rather than a subscription.
Work with Zapwiser →Build the whole thing on Creator and see whether it works before any money changes hands. You upgrade at the point it stops being an experiment.
A workflow extracts the terms from every proposal, document or dataset, standardises them and ranks the options — collapsing days of manual cross-checking into minutes.
The routine reading and checking is handled and traced automatically, so your most experienced people spend their time on the calls that actually need judgement.
Every result carries its inputs, assumptions and steps — which is what you need when a client, an auditor or a board asks why.
Every plan includes the visual canvas, AI models, Python and your own model API key. These are the things that differ.
| Capability | Creator (Free) | Private / Business |
|---|---|---|
| Visual canvas, agents, Python and APIs | ✓ | ✓ |
| Bring your own model API key | ✓ | ✓ |
| Scheduled runs and webhooks | ✓ | ✓ |
| File upload size | 5 MB | Sized to your needs |
| Python libraries | Built-in set | Your own packages |
| Compute | Shared | Dedicated virtual environment |
| Deployment | QMerse cloud | Cloud, private cloud or on-premises |
| Custom integration nodes & connectors | — | ✓ |
| Shared workspaces & collaboration | — | ✓ |
| Support | Community | Priority, with a named contact |
Short version: Creator has no time limit. You talk to us when you outgrow the shared environment.
No. Creator is free forever, subject to its file size, library and compute limits. It is not a trial and we do not ask for a card.
Three that matter: individual file uploads are capped at 5 MB, you have the built-in Python libraries rather than arbitrary packages, and your workflows run on shared compute alongside everyone else’s. Everything else — the canvas, agents, scheduling, your own model key — is the same product we run for paying customers.
Yes. Individuals, consultants and companies are all welcome to build on Creator, including for real client work. We would rather you proved the thing works before anyone talks about money.
When you hit one of the three Creator limits, or when the answer to “where does this run?” stops being “wherever is fine”. In practice that means bigger files, your own Python packages, dedicated compute you are not sharing, a connector to a system we do not support yet, or the workflow running inside your own infrastructure.
Connect your own model API key — OpenAI, Anthropic, Azure or a local model. You then control which model runs and what it costs, and we do not mark up inference.
Because the things people upgrade for — dedicated compute, your own packages, a custom connector, a deployment in your environment — are all specified rather than bought off a page. A short conversation gets you sized properly instead of a plan you have to work around. We would rather have two honest options than six invented ones.