Skip to content

Models and plans

Which AI models each plan can use, how the monthly allowance is counted, and what changes when you bring your own key.

1 min readUpdated

Every plan comes with Inkwell-hosted AI. Plans differ in two ways — how much you can run each month, and which models you can choose.

The allowanceLink to this section

Skill runs are counted per month, ghost-text completions and inline edits per day. The limits for every plan are on Plans and quotas; Settings → Plan & usage shows how much of this period's allowance you have used.

Model choiceLink to this section

Paid plans unlock the more capable models for skills and chat; the model picker in a skill's options, and in the Chat tab on Ask, lists what your plan can use. The free plan runs the standard model. Which model a plan offers is on the pricing page.

Your own keyLink to this section

With your own Anthropic or OpenAI key, the model list is the provider's list, and runs draw on your provider account rather than the Inkwell allowance. See Bring your own key.

Running outLink to this section

When the month's skill allowance is used up, a run stops before it starts and the Skills tab says so, with the options: wait for the reset, add credits, or use your own key. See Credits and overage.

What next

  • Plans and quotas

    Every plan's limits — projects, draft lines, save points, skill runs, completions, collaborators, share links, storage, and export formats.

    Reference · 1 min read

  • Bring your own API key

    Use your own Anthropic or OpenAI key instead of Inkwell's hosted AI, pick the model, and see how a key changes what counts against your plan.

    How-to · 1 min read

  • Credits and overage

    What happens when the month's AI allowance runs out, how prepaid credit packs work, and how turning on overage lets runs draw them down instead of stopping.

    Concept · 1 min read

What writers have asked about this — open the request to follow it.