Costs and Spending

Both providers work the same way at the top level: you put money in, and it is drawn down as you use it. There is no subscription, no monthly minimum, and nothing is charged if you do not use the tools.

Usage-based billing worries people who have never used it, almost always because of one fear: waking up to a bill they did not expect. The structure makes that very hard to happen.

The balance is prepaid

This is the most important thing on the page. You are not on an account with a monthly invoice that arrives after the fact: you add credit first, and the tools draw it down. When the balance reaches zero the features stop working until you top up again.

The most you can spend is the amount you have already put in. Start with a small balance while you learn what your usage looks like.

Language: priced per token

The Assistant, Flow Studio AI and the ✦ Smart mode of File Search are billed by the token, a chunk of text roughly the size of a short word or part of one.

Two things follow from that. First, you pay for the answer as well as the question, and a long answer costs more than a short one. Second, a conversation is charged cumulatively: each new message carries the earlier turns along as context, so a long thread costs more per message than a fresh one. Starting a new conversation when you change subject is the cheapest habit there is.

Prices vary enormously between models: the range between a small open model and a frontier one is not a few percent, it is orders of magnitude. Smart search always uses a small, fast model, so a search costs very little.

Images, video and 3D: priced per generation

Generative AI Studio is billed per output. An image has a fixed price for a given model and resolution; video is more expensive and usually scales with duration; 3D reconstruction is its own rate.

Here the number of attempts is what drives the bill. Refining a prompt through ten variations costs ten generations, so it pays to be specific in the prompt rather than to iterate blindly.

Where to see the actual prices

Both providers publish a full price list per model, and it is the authoritative source, rates change as models are released and retired:

• OpenRouter: openrouter.ai/models shows the per-token price of every model alongside its capabilities.
• fal.ai: fal.ai/models lists each image, video and 3D model with its per-generation cost.

See it, and cap it, in RhinoArtisan

The Assistant panel shows what the current conversation has cost so far, next to the Auto button, as OpenRouter bills it; hover it for today’s total. In Options → Artisan → Assistant you can set a daily limit in dollars: once the Assistant’s calls reach it, the Assistant stops for the day and says so in the chat.

The Assistant also keeps its costs down on its own. With Claude models, the long instructions and the tool descriptions that every message repeats are cached by the provider for a few minutes, so the second and later calls of a conversation pay a fraction for them. And each message is first read by a small, fast model that decides which tools it needs, so a short question doesn’t carry the whole toolbox.

Set limits at the provider

Both dashboards let you go further than the balance itself:

• Spending limits: cap how much a key may consume over a period.
• Per-key separation: issue one key per machine or per user, so you can see who is spending what and revoke one without disturbing the others.
• Usage history: a breakdown by model and by day, which is the honest answer to “where is the money going”.

Choose models deliberately

The gap between a cheap model and a frontier one is not a rounding error, it is orders of magnitude. Which means the model dropdowns in Assistant and Integrations options are the biggest cost lever you have, bigger than any change in how much you use the tools. See Choosing models for how to pick.

Habits that cost less

• Start a new conversation when you change subject. Every message in a thread carries the earlier turns as context, so long threads get more expensive per message. A fresh conversation resets that.
• Explore on a cheap image model, finish on an expensive one. Most of your generations are drafts you will discard. Pay the top rate only for the keeper.
• Write the prompt properly the first time. Ten vague attempts cost ten generations. One specific prompt often costs one.
• Ask the Assistant precise questions. A focused question gets a short answer, and you pay for the answer too.

In a studio with several seats

Give each machine its own key rather than sharing one. You get per-seat usage figures, you can set different limits, and if someone leaves you revoke a single key instead of rotating one that everybody is using.

Nothing is spent in the background

RhinoArtisan does not make AI calls on its own. Every charge corresponds to something you asked for: a message sent, a script generated, an image produced, a Smart search run. Closing the panel spends nothing, and leaving RhinoArtisan open overnight spends nothing.