WordPress AI Cost Calculator: What Will WordPress 7.0's AI Features Really Cost You?
Since WordPress 7.0, one AI provider key (Settings → Connectors) powers every plugin on your site — and core gives you no estimate of what that will cost before the invoice arrives. This calculator answers the question up front: pick the AI features you run (or plan to run), and see the estimated monthly bill, per feature and in total.
Prices are June-2026 published list prices per million tokens — the same bundled rate table our free AI Spend Monitor plugin uses. Estimates err high (provider caching discounts aren’t modeled), which is the direction you want an estimate to err.
How to read the numbers
Three things decide an AI feature’s cost: how often it runs, how much context it sends (prompt tokens — usually far more than people expect, because system prompts, page content, and chat history all count), and which model answers. The presets above use typical token counts we’ve observed; your plugins may differ — which is the point of measuring instead of guessing.
A few patterns worth knowing:
Chat is the budget eater. A support chatbot doing 40 replies a day at ~800 prompt tokens each runs real money on a premium model — and chat history makes prompts grow through a conversation. If one feature deserves a cheap model (Gemini Flash, GPT-5 nano), it’s chat.
Background features run without you watching. Auto-generated alt text, scheduled summaries, “AI-enhanced” cron jobs — these never show you a spinner, so the first sign of volume is the invoice. This category caused most of the surprise bills we’ve seen discussed.
Model choice swings cost ~100×. The same alt-text job costs two orders of magnitude more on Claude Opus than on Gemini 2.0 Flash. WordPress 7.0 lets every plugin pick its own model against your shared key — you may not know which model a plugin chose until you look.
From estimate to actual
This calculator tells you what your AI features should cost. What they actually cost — per plugin, per model, per day — is what the free AI Spend Monitor records on your own site: exact token counts (they match WordPress core’s request logs line for line), estimated cost per plugin, a 30-day chart, and CSV export. No account, zero outbound requests.
Download AI Spend Monitor free →
And if the numbers above made you wince: AI Spend Governance adds per-plugin monthly budgets and a hard stop that blocks calls before your provider charges anything.
Prices: published list rates per 1M tokens, reviewed June 2026, standard (non-batch, non-cached) tier. Providers change prices — treat your provider’s pricing page as authoritative, and your invoice as the source of truth.