Credits

Prompt2CAD bills in credits. One credit is one cent of "list price." Everything else is just arithmetic.

Fixed prices

Action	Cost
Generation (new model from a prompt — first turn)	30 credits
Iteration (refining an existing model — follow-up turns)	15 credits
Render (photorealistic image)	50 credits
Continuation (multi-step tool round-trip within one turn)	0
Parameter slider drag	0
PNG screenshot	0
Project switching, viewport rotation, anything UI	0

These prices are fixed. No matter how many tokens a generation actually consumes — whether your prompt is two lines or 200, whether the agent thinks for ten seconds or three minutes — you pay 30 credits per generation. Same for iterations and renders.

That's intentional. It means you can iterate without doing mental math, and it shields you from the agent's internal cost variance.

The free trial

Every new account gets 160 free credits on first sign-in. AI chat is charged by actual token usage, so the exact number of actions varies with the complexity and length of the conversation. As a guide, a new design is typically around 100 credits, a refinement around 50 credits, and a render costs 50 credits.

The trial is intended to let you create a first model and test conversational or parametric refinement before topping up.

When you "have to pay"

Two gates:

1. Insufficient balance. If your balance drops below the cost of the next action, the chat input shows an "out of credits" banner and the action is blocked. Top up to continue.
2. CAD / 3D exports. STEP, DXF, GLB, DAE, OBJ, STL exports are gated until your lifetime credited amount exceeds the free trial. In practical terms: as soon as you top up once, the gate opens forever. PNG export is always free.

The two gates are independent. You can have a healthy balance (1,000+ credits) and still hit the export gate if you've never topped up — those credits came from the trial.

What counts as a generation vs an iteration

The agent classifies each chat turn:

• Generation — you're building from nothing, or you've made changes that require rewriting the geometry from scratch.
• Iteration — you're refining an existing model. The agent makes a targeted change rather than rebuilding from scratch.
• Continuation — the agent is finishing a multi-step action it started (e.g. it took a screenshot and is now interpreting it). Free.

You don't have to think about which is which — the agent picks. But it explains why some turns cost 30 and others 15.

Failed turns

If a chat turn errors out before producing usable output — a model timeout, a geo-block, a malformed reply — no credits are charged. The failed turn is still logged so we can investigate, but your balance is untouched.

If you suspect you were charged for a failed turn, email support with the project name and approximate time. Usage records carry timestamps and can be cross-checked.

Checking your balance

• Header chip — top right of /app. Always visible.
• Credit runway banner — appears in the chat when your balance is close to running out, with the count of remaining generations/iterations.
• Per-action confirmation — Render explicitly confirms "50 credits" before charging.

Cost transparency

The actual cost of an action — to us, in dollars paid to the model provider — is tracked internally but is not what you pay. We charge fixed credits; we absorb the variance. This means our gross margin per call varies; yours does not.

Why we picked this model

The alternative is per-token billing (you pay for what the agent actually consumes). It's "fair" in the sense that a short turn costs less than a long one, but it makes the cost of iterating unpredictable. Fixed-credit billing lets you say "I'll spend 200 credits on this design," budget it, and stick to it.

Next: Top-up — how to actually pay.