Plus and Pro

After the top-up, check that it’s active

After you submit at cz.2api.dev:

1. Refresh chatgpt.com, or fully quit the app and open it again.

2. Click your avatar in the bottom-left corner to open the profile menu. Make sure you’re signed in to the account you submitted for the top-up.

3. Open settings and check your current plan under Billing. If it shows Plus or Pro, it’s active. You can also open the Codex usage page, which shows the usage limits for your plan.

It still shows Free:

• Make sure the account and the sign-in method match the ones you used for the top-up. See the pitfalls in Sign in.
• Sign out and sign in again. If you’re signed in on several devices, sign in again on each one.
• Check again in a few minutes.
• If it still doesn’t work, contact support with your account email and the time of the top-up.

Once it’s active: download the app, sign in with the account you topped up, then read Using ChatGPT and Codex.

What Plus and Pro include

The table below follows OpenAI’s English pricing page. Pro comes in three tiers, which differ mainly in usage.

	Plus	Pro
Models	GPT-6 reasoning models	Everything in Plus, plus Pro reasoning based on GPT-6 Astra
Codex and Work	Expanded usage	Longer Codex and Work sessions
Image generation	More complex, more accurate image generation	Unlimited and faster
Deep research	Expanded	Highest tier
Memory and context	Expanded	Highest tier
Projects, scheduled tasks, custom GPTs	Included	Expanded
Other	Early access to new features	Research previews of new features, the always-on assistant Dot, monthly API credits (the amount depends on the tier)

Context length: on Plus, 54K for instant models and 256K for reasoning models. On Pro, 128K and 400K.

Features listed as “unlimited” still have limits against abuse. Go by what OpenAI says.

Usage limits

• Chat: text chat has no message cap on any plan, but there are limits against abuse.

• Codex and Work: Plus counts usage in 5-hour windows, and there may also be a weekly limit. Pro currently has no 5-hour limit. Local tasks and cloud tasks share the same usage, and cloud tasks use it up faster. Generating images uses it 3 to 5 times faster than ordinary conversations.

• Below is roughly how many local messages Plus can send every 5 hours. These are OpenAI’s estimates, not fixed caps, and they depend a lot on the size of the task:

  Model	Every 5 hours
  GPT-6 Astra	5–45 messages
  GPT-6.1 Sol	15–160 messages
  GPT-6.1 Terra	35–285 messages
  GPT-6 Sol	15–150 messages
  GPT-6 Luna	350–3,000 messages

• Check remaining usage: open chatgpt.com/codex/settings/usage to see your current limits and reset times. You can also type /status or /usage in the Codex CLI.

• Used it all up: wait for the reset, or buy extra credits in ChatGPT to keep going. You don’t need to upgrade your plan.

Run Codex on your subscription

Once your plan is active, choose Sign in with ChatGPT in Codex. Codex then uses the usage included in your subscription. See Sign in.

If your subscription’s allowance runs out and you have 2api balance, connect Codex to magpie and pick a model marked · 2api from Codex’s model menu. See Use it with a ChatGPT subscription.

Free and Go can only use Codex (GPT-6 Luna) in the desktop app. Plus and Pro can also use it in the CLI, the IDE extension, on the web and on iOS.

---

This page follows OpenAI’s official documentation, checked in October 2026. OpenAI updates often, so button names and menu locations may differ slightly. Go by what you see on screen.