MyJackfruit

Help
Browse topics · AI costs and limits

AI costs and limits

← AI costs and limits

How do I see what my AI use has cost so far, and why was one day so much?

Short answer

Your AI's maker shows it: in Claude Code type /usage, or open Settings > Usage at claude.ai; GilbApp does not count AI use.

Before you start

Words used here (1)
GilbApp
GilbApp is the app, opened in a web browser at https://gilb.com/app, where you open, fill and share your vaults.
Deeper

Also asked:

How do I see what I have spent so far? Your AI's maker shows it, not GilbApp: in Claude Code type /usage, or open Settings > Usage at claude.ai; for Codex, see OpenAI's usage page. If you turned on usage credits, /usage also shows this month's spend on them.

How do I see how much of my AI plan the agents have used? In Claude Code type /usage: it shows plan usage bars and what used the most, such as helper agents; press d or w for the last 24 hours or the last 7 days.

Your AI is your own, so for your AI you pay its maker, Anthropic, OpenAI or xAI, not GilbPlatform. GilbApp counts no AI use and shows no cost: in its chat, each answer is only marked with the AI that gave it and your Claude subscription or your SuperGrok. The answer What will my AI cost while I plan and build, and can I set a limit? tells what you pay.

Claude. With a Claude plan, Pro, Max or Ultra, you pay a set price each month, and the plan has usage limits, on most plans every 5 hours and every week. At claude.ai, under Settings > Usage, Anthropic says You can see how much you've used and the time remaining under Current session; the same page shows your next weekly reset. In Claude Code, Anthropic's AI that you run in a terminal, the window where you type commands, type /usage and press Return: subscribers see plan usage bars, activity stats, and a usage breakdown on the same screen. The breakdown shows what used the most, such as helper agents, with suggested changes; press d or w for the last 24 hours or the last 7 days. Anthropic says those figures "are approximate and computed from local session history on this machine, so usage from other devices or claude.ai is not included." The dollar figure in its Session part is for paying by use and does not count for a plan. If you turned on usage credits, Anthropic's pay-as-you-go use beyond a plan's limits, /usage also shows this month's spend on them (https://code.claude.com/docs/en/costs, https://support.claude.com/en/articles/11647753-understanding-usage-and-length-limits, read 13 November 2026).

Codex, OpenAI's AI for writing code. OpenAI shows your limits and when they reset on its usage page, https://chatgpt.com/codex/settings/usage, and in Codex you can type /status to see your remaining limits during an active Codex CLI session (https://learn.chatgpt.com/docs/pricing, read 13 November 2026).

Grok. xAI sets SuperGrok's limits and shows use on its own pages; we could not read them, so this answer says nothing about them. If you need help finding it, write to [email protected].

Why one day used so much. Building uses far more than chatting. Anthropic says: "each Claude Code turn carries file contents, tool calls, and multi-step reasoning, so one debugging session can consume more than a day of chat." It also says Claude Code sends your full conversation with every request, so a session left open all day costs more with each message; your first message after a long break reads the whole conversation again; and every subagent, a helper agent your AI starts, sends its own requests on top of the main conversation's. While it builds, your AI starts such helpers: Agent.Coder, the agent that writes your app's code, has each built step checked by a separate agent. A bigger model, the version of Claude you choose with /model, also uses more: Anthropic says Sonnet costs less than Opus. Starting fresh with /clear costs nothing, and the /usage breakdown suggests changes for whatever used 10% or more (https://code.claude.com/docs/en/costs, read 13 November 2026).

Your plan's own budget. If your plan names a budget, such as money or hours, Kai-Zen writes what each delivered step used in the plan; the answer How do I track estimates and actuals? shows where. If you have reached a limit, see My AI says I have reached my usage limit. What now?

Anthropic, OpenAI and xAI set their plans, limits and prices, and may change them.

This answer is written for a computer screen.

Source: Anthropic, read 13 November 2026: https://code.claude.com/docs/en/costs (/usage: Session block intended for API users, subscribers see plan usage bars, activity stats, and a usage breakdown, d or w for 24 hours or 7 days, approximate and computed from local session history on this machine, usage-credits row; Why usage climbs in a long session: full conversation with every request, cache misses after a break, subagents; Enterprise guide sentence on a Claude Code turn; Sonnet costs less than Opus; /clear costs nothing), https://support.claude.com/en/articles/11647753-understanding-usage-and-length-limits (Settings > Usage, Current session, weekly reset). OpenAI, read 13 November 2026: https://learn.chatgpt.com/docs/pricing (usage page, /status). xAI: no page could be read. GilbApp 49b1ae53 (https://gilb.com/app/version.json, read 13 November 2026): src/components/agent/AgentChat.tsx:1215-1227 (your Claude subscription, your SuperGrok); no count of AI use in src/components/agent. Standard Agent.Coder.AfterIncrement (Agent.Tester started as a separate agent), read 13 November 2026. Live help pages, sentences shared: What will my AI cost while I plan and build, and can I set a limit? (pay its maker; plan limits every 5 hours and every week; usage credits gloss; /clear); How do I track estimates and actuals?

Checked against GilbApp d529620.