When does the weekly limit reset?
Normally both refill once a week. After a reset they differ: on Codex, the next weekly reset is 7 days after your first message following the reset, so send one right away; on Claude, the weekly reset time stays fixed.
Read article →When does the 5-hour limit reset?
Five hours after your first message. For the exact time, check /usage or Settings › Usage on Claude, and the usage page or /status in the CLI on Codex. Codex removed its 5-hour limit in July 2026 and brought it back for Plus only in late August; Pro has none for now.
Read article →Why did my weekly limit reset early?
Most likely the vendor reset everyone. Match the date against the reset log on whenreset to see which one it was. On Codex, the next weekly reset is 7 days after your first message following the reset, so send one right away. On Claude, your weekly reset time does not move, so the reset is a free extra.
Read article →How do Codex banked resets work, and when do they expire?
Open the usage page in settings, select the line that shows an available reset, and confirm: the 5-hour limit and the weekly limit refresh together. /usage in the CLI works too. A banked reset usually expires 30 days after it is granted; go by the expiration shown in your account, because an expired one is not reissued. Send a message right after using it: the next weekly reset is 7 days from that message.
Read article →How do I check how much Codex or Claude usage I have left?
Claude: type /usage in Claude Code, or open Settings › Usage on claude.ai. Codex: open chatgpt.com/codex/settings/usage, or type /status in the CLI or the VS Code extension. Each one shows how much you have used and when it resets. whenreset tracks the resets a vendor gives everyone, not your own usage.
Read article →How do you save tokens in Claude Code and Codex?
Three things use the most tokens: a session that has been open for hours, subagents, and the fixed content sent with every turn (CLAUDE.md or AGENTS.md, MCP servers, skill descriptions). Start with three steps: clear the chat when you switch tasks, keep reasoning effort at medium (the Codex default; /effort medium in Claude Code), and give simple subagents a smaller model. Tokens are what the 5-hour and weekly limits are measured in.
Read article →Long sessions use more tokens: /compact or /clear?
Use /clear when you switch tasks and /compact only in the middle of one. Every message re-reads the whole conversation, so the longer the session, the more each line costs. /clear costs nothing; /compact is a large request in itself.
Read article →Subagents use too many tokens: how do you give them a cheaper model?
In Claude Code, write model: haiku at the top of the subagent file. In Codex, set default_subagent_model under [agents] in ~/.codex/config.toml. Each subagent runs its own requests against the same allowance, so searching, reading files and similar simple work is cheapest on a small model.
Read article →Which reasoning effort and model use the fewest tokens?
Use the default effort for everyday work (Medium on Codex, high on Claude Code) and raise it only for hard problems. When tokens are tight, drop one level and one model size. Higher effort is slower, uses more tokens and does not guarantee a better result.
Read article →How many tokens do CLAUDE.md and AGENTS.md use, and how long should they be?
There is no fixed number: the file costs what its length costs, and it rides along with every message in the session. Anthropic suggests keeping each CLAUDE.md under 200 lines. Codex stops adding AGENTS.md files once they reach 32 KiB combined. Keep the rules you need every time and move the rest out.
Read article →How many tokens do MCP servers use, and how do you cut it?
By default Claude Code puts only MCP tool names and server instructions into context and loads a full tool definition when it is used, so a few more servers cost little. On Codex, every MCP server adds context to every message. The fix is the same on both: turn off what you do not use.
Read article →How many tokens do skills and plugins use each turn?
Each turn carries only the name and description of every skill, not the full text; one plugin with 6 skills measured about 221 tokens per turn. The full skill loads when it is used. Claude Code lists what each plugin adds per turn in /usage. Codex works the same way but shows no per-skill number.
Read article →Do token-saving tools like rtk and caveman really work?
A little, and far less than advertised. JetBrains ran paired tests on Claude Code and compared the bill: caveman advertised 65% and cut output tokens by 8.5%; rtk advertised 60–90% and cost 7.6% more at low reasoning effort, the same at high. Neither made task quality worse. Before you install one, use /usage to see where your tokens go; afterwards, read your usage, not the number the tool reports.
Read article →Limit almost used up, days before the reset: what to do?
Check two things first: what is left and when it resets (/usage in Claude Code, /status in Codex). Then cut back, fastest effect first: fewer subagents, lower reasoning effort, a smaller model, a new session for each new task. If that is not enough, the official ways out are to wait for the reset, upgrade your plan or buy extra usage; on Codex you can also use a banked reset.
Read article →Codex says "Selected model is at capacity". How do I fix it?
Check service status, distinguish a capacity error from a usage limit, and try a supported alternative before contacting support. The error alone does not prove an account restriction.
Read article →
Usage wiki: when limits reset and how to save tokens
When limits reset and how to make them last longer. Browse by topic, with separate answers for Codex and Claude, based on official docs with quotes and links.