Skip to content
Codex DesignBETA

Codex pricing: plans, usage limits, and credits.

Codex is included in every ChatGPT plan, from Free to Enterprise, so you don’t buy it separately. What changes between plans is how much you can use it, which models you get, and whether you can buy more. This guide summarizes OpenAI’s Codex pricing page as of November 2026 and explains what to do when you hit a limit.

Updated November 6, 2026 · 6 min read

Is Codex free?

Yes. Codex is included in ChatGPT Free, Go, Plus, Pro, Business, Edu, and Enterprise. OpenAI describes the Free plan as a way to explore Codex on quick coding tasks; it includes GPT-6 Luna at standard speed in the desktop app, subject to rollout. Image generation isn’t available on Free. For regular work, most people need Plus or above.

Codex prices by plan

PlanPriceWhat OpenAI says it’s for
Free$0Exploring Codex on quick coding tasks
Go$8 per monthLightweight coding tasks
Plus$20 per monthA few focused coding sessions each week
Pro$100, $200, or $500 per monthNo five-hour limit for now; Ultrafast on $500
Business$20 per user per month billed annually (2+ users), $25 monthlyTeams, with larger cloud machines
Enterprise and EduContact salesOrganizations, with priority processing

Prices are in US dollars and come from OpenAI’s Codex pricing page. Check it before you buy, because limits and models change often.

How usage limits work

Codex limits are counted in messages per five-hour window, and weekly limits may also apply. How many messages you get depends on the model: a light model goes much further than the most capable one. OpenAI’s estimates for Plus and standard Business seats:

ModelLocal messages per 5 hours
GPT-6 Astra5–45
GPT-6.1 Sol15–160
GPT-6 Sol15–150
GPT-6.1 Terra35–285
GPT-6 Luna350–3,000
  • Pro plans currently have no five-hour limit.
  • Enterprise and Edu with flexible pricing have no fixed rate limits; without it, each seat has the same limits as Plus.
  • Local and cloud work share one allowance. Cloud tasks may use more of it than local messages.
  • Fast mode uses 2.5× your included usage. Ultrafast mode, on Pro $500 for GPT-6 Astra and GPT-6.1 Sol, uses 8×.

Your current limits and when they reset are on your Codex usage dashboard.

Credits: paying for more

Plus and Pro users can buy extra ChatGPT credits when they run out. Business, Enterprise, and Edu workspaces with flexible pricing buy workspace credits. Credits are charged per million tokens at a rate that depends on the model, so the same task costs far fewer credits on GPT-6 Luna than on GPT-6 Astra. OpenAI says a typical GPT-5.6 Sol task uses about 2–15 credits. The price of a credit depends on your plan.

Using Codex with an API key

You can also sign in to the Codex CLI, SDK, or IDE extension with an API key and pay standard API prices instead of using a plan. Cloud features such as GitHub code review and Slack aren’t available that way, and the models you get follow your API key.

What to do when you hit the limit

  1. Switch to a lighter model for routine work, such as GPT-6 Luna, and keep the strongest model for hard problems.
  2. Turn off fast mode. It uses your allowance 2.5 times faster.
  3. Prefer local tasks over cloud tasks when either works.
  4. Buy credits on Plus or Pro to keep going without waiting.
  5. Wait for the window to reset, or move to a higher plan if you hit the limit every week.

For UI work, a lot of usage goes into regenerating screens. Asking for two or three directions at once and comparing them side by side, instead of rerunning one prompt again and again, gets you to a design you like in fewer messages.

Frequently asked questions

How much does Codex cost per month?

Nothing on ChatGPT Free, $8 on Go, $20 on Plus, and $100, $200, or $500 on Pro. Business is $20 per user per month billed annually, or $25 monthly.

Is Codex included in ChatGPT Plus?

Yes. Plus includes Codex in the desktop app, the web, the CLI, and the IDE extension, with limits OpenAI describes as enough for a few focused coding sessions a week.

When do Codex limits reset?

Limits are counted per five-hour window, and weekly limits may also apply. Your usage dashboard shows the exact reset time.

Keep reading

An independent plugin for Codex. Not affiliated with or endorsed by OpenAI.