Hit your Claude Code limit? 4 ways to keep working

Brandon Omey — Mon, 28 Sep 2026 06:06:48 +0000

Claude Code just told you you've reached your usage limit, halfway through a build. You have four real options: wait, switch accounts, switch agents, or pay per use. Here's what each costs, whether your conversation comes along, and which one we built into Technically Not.

First: how Claude plan limits work

If you use Claude Code with a Claude Pro or Max plan, you have two limits at once.

A session limit. It resets every five hours.
A weekly limit. It resets on a fixed schedule that Anthropic sets for your account.
One shared pool. Chatting in the Claude app and coding in Claude Code count against the same limits.

How far a session goes depends on the model, the size of your project and how long the conversation has run. Bigger models such as Opus use more of it than Sonnet or Haiku. In Claude Code, /status shows where you stand, /model switches to a lighter model, and /compact shrinks a long conversation so each message costs less.

1. Wait for the reset

The simplest option. The limit message says when your usage comes back. If it's the five-hour session limit, that's a coffee break or an evening. If it's the weekly limit, it can be days.

Costs: nothing extra.
Conversation: stays exactly where it was. Type "continue" when the time comes.
Catch: you have to be there when it resets, or remember to come back.

Some plans also get an occasional limit reset from Anthropic: under Settings › Usage, "Reset for free" sets your session or weekly limit back to full once. Check there before you wait.

In Technically Not: Auto-continue (on by default, Settings › Agents) marks the agent "Paused · usage limit" and stops its timer. About a minute after the reset time shown in Claude Code's message, it tells the agent to carry on without repeating finished work. You can walk away.

2. Switch to another account

If you have a second Claude account you're allowed to use (a work account and a personal one, say), each has its own limits.

By hand: in Claude Code, run /login and sign in with the other account. Your conversations are saved on your computer, so claude --continue in the same project folder opens the latest one again. Power users keep both accounts signed in at once by giving each terminal its own settings folder with CLAUDE_CONFIG_DIR. It works, and it's free, but you're retyping commands mid-build.

In Technically Not: add each account once under Settings › Accounts; each gets its own folder and stays signed in. When a terminal panel shows Claude Code's limit message, the agent's card says so and offers a Switch button. The agent restarts on the other account and resumes the same conversation when the CLI can, and tells you whether it resumed or started fresh. Codex accounts work the same way. See how account switching works.

Costs: the other account's plan (Pro is $20/month in the US; Max is $100 or $200/month).
Conversation: comes along, as long as it was saved.
Catch: it doesn't add usage to any account. Each account keeps its own limits and Anthropic's terms still apply.

3. Keep going with another agent

Claude Code isn't the only coding agent. Codex, from OpenAI, is included with ChatGPT plans (Plus is $20/month, Pro starts at $100/month), and it has its own separate limits. When Claude is out, Codex can often pick up the next task in the same project folder.

Costs: nothing extra if you already pay for ChatGPT.
Conversation: doesn't carry over. Codex can't open Claude Code's conversation, so give it a short summary: what you're building, what's done, what's next. It can read the files Claude already wrote.
Catch: a different agent works a little differently, and Codex has limits of its own.

In Technically Not: Claude Code and Codex run side by side in terminal panels in one window, on the same project, each with its live status. How to run Claude Code and Codex together.

4. Pay per use, or upgrade

If you'd rather not stop at all, you can pay for more.

Usage credits. On Pro and Max, turn them on under Settings › Usage. After your included usage runs out, you keep working and pay standard API rates from a balance you top up. They cover Claude Code too, and you can set a monthly spending cap.
An API key. Sign Claude Code in with a Claude Console account instead of your plan, and you pay standard API rates for every message, with no plan limit.
A bigger plan. Max 5x ($100/month) or Max 20x ($200/month) gives you 5 or 20 times Pro's usage per session.
Conversation: credits and upgrades keep you in the same conversation. With an API key, it's a different sign-in, so continue the saved conversation the same way as in option 2.
Catch: pay-per-use has no natural stop. Set a cap before a long session. In Technically Not, add the key as an API-key account (Settings › Accounts) and switch to it like any other account when your plan runs out; the key is kept encrypted on your own computer and never sent to Technically Not's server.

Which one should you pick?

Hitting limits once in a while? Wait, and let the reset come to you. Hitting them every day on one account? A bigger plan is the tidiest fix. Already paying for a second account or for ChatGPT? That's the usage Technically Not helps you reach without breaking your flow: switch accounts in one click, or put Codex on the next task next to Claude. Read the account-switching page.

Technically Not is $15/month or $144/year (Founder: $99/year until December 25, 2026). No free plan and no trial, but the live demo shows the real interface first. It runs on Windows 10/11 and Linux, and you keep paying for your own AI plans.

Questions

How long until my Claude Code limit resets?

On the Pro and Max plans, the session limit resets every five hours. There is also a weekly limit that resets on a fixed schedule set for your account. Claude Code's limit message tells you when it resets.

Do Claude and Claude Code share the same limit?

Yes. On Pro and Max, everything you do in the Claude app and in Claude Code counts against the same usage limits.

Can I keep my conversation when I switch accounts?

Usually. Claude Code saves conversations on your computer, so after signing in with another account you can pick the conversation up again with claude --continue. Technically Not does this for you when the CLI can resume, and tells you if it had to start fresh.

Does Technically Not give me more usage?

No. Each account keeps its own plan and its own limits, and each provider's terms still apply. Technically Not makes moving between accounts and agents you already have quicker, and can continue the agent after the reset.

What does Technically Not cost?

Pro is $15/month or $144/year, and the Founder price is $99/year until December 25, 2026. There is no free plan and no trial, but you can try the real interface in the browser before you buy. You keep paying for your own Claude and ChatGPT plans.

Sources, checked 2026-09-27

Claude Help Center: What is the Pro plan? and What is the Max plan? (prices, five-hour and weekly limits)
Claude Help Center: Use Claude Code with your Pro or Max plan (shared limits, /login, /status, Console billing)
Claude Help Center: Manage usage credits for paid Claude plans and What is a limit reset?
Claude Help Center: Models, usage, and limits in Claude Code (/model, /compact)
Claude Code docs: CLI reference (--continue, --resume)
OpenAI: Codex pricing (plans that include Codex, limits, credits)

Technically Not is an independent app. It isn't made by, or affiliated with, Anthropic or OpenAI.

Disclosure: I make Technically Not, the desktop app mentioned above. The other three ways work without it.

DEV Community: Brandon Omey