What happens when you hit usage limits?

Active turns can finish subject to fair use. Plus and Pro users can buy additional credits to continue without upgrading. Business/Edu/Enterprise with flexible pricing can buy workspace credits. All users can run extra local chats via API key at standard API rates.

How does image generation count toward usage limits?

Image generation counts toward the same general usage limits and uses them ~3-5x faster than similar non-image turns; after limits it draws from credits. Not available on Free; API key users pay API pricing instead.

Where can I see my current usage limits?

Check the usage dashboard in ChatGPT, or use /status inside an active Codex CLI session. The page recommends checking every week or two.

What are tokens and credits?

Tokens measure what ChatGPT reads and writes; credits translate that into a simpler unit. After included limits, available credits let you keep working. Credit cost varies by model, context, reasoning, and tools.

What counts as Code Review usage?

Only reviews Codex runs through GitHub (e.g., @Codex tagged on a PR or automatic repo reviews). Reviews run locally or outside GitHub count toward general usage limits.

Can I extend my usage limits?

The page lists tips: shrink prompts, limit source material, match output to need, reduce AGENTS.md size, limit MCP servers, and switch to a smaller model (GPT-5.6 Terra or Luna) for routine tasks.

Codex Pricing in 2026: Plans, Limits and Real Costs

Plus at $20/month is the right tier for solo developers running a few focused coding sessions a week. Pro 5x at $100/month is the upgrade once Plus limits bite. Skip Free for real work; skip API Key unless you are wiring Codex into CI.

Verified 2026-08-21 · official pricing page · Codex in the directory

Quick pricing verdict

Starting price: $0 /month on the Free plan
Free plan: Free exists at $0 /month and is described as 'explore Codex capabilities on quick coding tasks.'
Best plan for most people: Plus at $20 /month, billed for 'a few focused coding sessions each week' with the GPT-5.6 model family.
Plan to avoid first: API Key, because it strips out cloud-based features like GitHub code review and Slack integration and bills per token.
Biggest hidden cost: Usage limits are not generous on Plus and Pro, and after you hit them you must purchase additional credits to keep working without upgrading.
Pricing checked: 2026-08-21, against the official page

Codex ships inside ChatGPT Free, Go, Plus, Pro, Business, Edu, and Enterprise, plus a token-billed API Key option for automation. ChatGPT Work and Codex share a usage pool, so Codex limits are also your ChatGPT Work limits.

Per-seat pricing is published for Free ($0), Go ($8), Plus ($20), Pro 5x ($100), and Pro 20x ($200). Business and Enterprise seat prices are not published on this page, and no annual billing figures are shown for any tier.

Plans compared

Plan	Price	Best for	Key limits
Free	$0 /month	Trying Codex on quick coding tasks	Local messages and cloud chats per 5h are not published in the limits table for Free. Voice on desktop is not available. Sites available only during public beta.
Go	$8 /month	Lightweight coding tasks	Specific per-model message limits are not published for Go on this page.
Plus	$20 /month	A few focused coding sessions each week, plus Codex on web, CLI, IDE extension, and iOS	Local messages per 5h: GPT-5.6 Sol 10-100, GPT-5.6 Terra 25-200, GPT-5.6 Luna 250-2,000, GPT-5.5 15-80, GPT-5.4 20-100, GPT-5.4 mini 60-350. Cloud chats and Code Reviews per 5h are 'Not available' for Plus in the table.
Pro 5x	$100 /month	Heavier daily coding than Plus can absorb; needed for Codex-Spark research preview	5x Plus Codex usage. GPT-5.3-Codex-Spark (research preview) included. Desktop voice ~1-2.5 hours per rolling 5h window. Specific per-model limits are not published in the table.
Pro 20x	$200 /month	Sustained heavy Codex work	20x Plus Codex usage. Unlimited ChatGPT Voice on desktop. Specific per-model limits are not published in the table.
Business / Enterprise	Not published	Teams needing SAML SSO, MFA, RBAC, SCIM, EKM, compliance and audit APIs, and Cloud-managed config	Enterprise/Edu on legacy pricing: ~45 minutes desktop voice. Business: ~45 minutes desktop voice. Legacy plans share Plus per-seat usage limits for most features; plans with flexible pricing scale by credits.
API Key	Pay only for the tokens Codex uses, based on API pricing	Automation in shared environments like CI	No cloud-based features (no GitHub code review, no Slack). Model availability follows API models available to your key. Local messages and cloud chats per 5h are 'Not available' in the limits table.

Source: Codex pricing page, checked 2026-08-21.

What each plan includes

Free

Includes: Access to Codex on web, CLI, IDE extension, and SDK/codex exec/scriptable workflows (per the feature availability table); entry into Codex capabilities on quick coding tasks.
Missing: No image generation. No ChatGPT Voice on desktop. No Codex cloud chats, no cloud environments, no GitHub code review, no Slack/Linear cloud integration. Sites only during public beta.
Avoid if: You plan to ship real features; Free is positioned as a trial.
Verdict: Useful only as a sandbox.

Go

Includes: Lightweight Codex usage at $8 /month.
Missing: Specific feature differences versus Plus are not spelled out on this page.
Avoid if: You want cloud-based integrations like automatic code review or Slack integration, or access to the full GPT-5.6 model family listed under Plus.
Verdict: A middle tier with limited published detail; Plus is a clearer value jump for $12 more.

Plus

Includes: Codex on web, CLI, IDE extension, and iOS; cloud-based automatic code review and Slack integration; GPT-5.6 family (Sol, Terra, Luna); Flexibly extend usage with ChatGPT credits; other ChatGPT features in the Plus plan; local code review with /review.
Missing: No Codex-Spark research preview. No Codex cloud chats or cloud environments. No GitHub issue/PR delegation with @codex. No Linear cloud integration. No SAML SSO, MFA, or SCIM. Desktop voice only ~15-30 minutes per 5h window.
Avoid if: You run long sessions, large projects, or need Codex-Spark; you will burn through Plus quickly.
Verdict: Best starting tier for a solo developer doing real work.

Pro 5x ($100)

Includes: Everything in Plus, plus GPT-5.3-Codex-Spark (research preview), 5x Codex usage vs Plus, ~1-2.5 hours desktop voice per 5h, unlimited ChatGPT Voice not included.
Missing: Cloud chats and code reviews per 5h are 'Not available' in the published table; cloud chats appear only for plans with cloud entitlements. No Business/Enterprise admin tooling.
Avoid if: You mainly need Slack or GitHub PR review automation; that is Plus territory.
Verdict: The upgrade when Plus limits feel restrictive.

Pro 20x ($200)

Includes: Everything in Pro 5x with 20x Plus Codex usage and unlimited ChatGPT Voice on desktop.
Missing: Tasks started through Voice still draw from your Codex usage budget, so 'unlimited voice' does not equal unlimited Codex work. Cloud chats/code reviews per 5h are 'Not available' in the table.
Avoid if: You would not actually use the 5x headroom; Pro 5x is cheaper.
Verdict: Only worth it if you consistently max out Pro 5x.

Business / Enterprise

Includes: Everything in Plus (and Pro features where applicable) plus SAML SSO, MFA, workspace RBAC and custom roles, SCIM/EKM/domain verification, enterprise retention and residency controls, no training on API or business data by default, analytics dashboard and API, compliance API and audit logs, Codex Security for connected GitHub repos, GitHub issue/PR delegation with @codex, GitHub code review, Slack and Linear cloud integrations, Codex cloud chats and cloud environments.
Missing: Seat pricing is not published. Voice on desktop caps at ~45 minutes for Business and Enterprise/Edu on legacy pricing (desktop voice costs ~6 credits/minute on credit-based billing).
Avoid if: You are a solo founder; this is overkill.
Verdict: Right choice only when admin, security, and team collaboration features matter.

API Key

Includes: Codex in CLI, SDK, or IDE extension. Local code review. Pay per token. Model availability follows API models available to your key.
Missing: No cloud-based features (GitHub code review, Slack). Voice on desktop is not available via API Key. Image generation uses API pricing, not included limits.
Avoid if: You want a turnkey coding assistant with web UI and cloud review.
Verdict: Only for CI, scripts, and self-hosted automation.

Add-ons, overages, and surprise costs

Cost item	Trigger	Published amount	Billing behaviour
Additional credits after limits	Reaching your plan's usage limit during normal Codex work	Not published	Plus and Pro users can purchase extra credits to continue without upgrading; Business/Edu/Enterprise with flexible pricing can buy workspace credits.
Image generation drain	Generating images during Codex turns	Not published	Image generation uses included limits ~3-5x faster than similar non-image turns, then draws from credits once limits are hit. Not available on Free.
Fast mode premium	Enabling Fast mode on supported models	Not published	Fast mode consumes credits at a higher rate; included limits are used up faster.
Voice on credit-based workspaces	Using ChatGPT Voice on Desktop on Business/Edu/Enterprise with credit-based or pay-as-you-go billing	~6 credits per minute	Voice has a separate plan-dependent allowance but still draws from Codex usage budget for tasks.
Codex-Spark demand-based limits	Using GPT-5.3-Codex-Spark (research preview, Pro only)	Not published	Runs on specialized low-latency hardware; usage is governed by a separate usage limit that may adjust based on demand.
Daybreak Red	Requesting Daybreak Red model access	312.5 credits / 1M input, 31.25 cached, 1875 output	Requires separate approval and provisioning; Daybreak Blue is gated by Trusted Access for Cyber.

Source: Platform pricing docs, checked 2026-08-21.

Which plan should you choose?

Solo developer shipping features in short weekly sessions: Plus ($20/month) — Matches 'a few focused coding sessions each week' and unlocks the GPT-5.6 family plus GitHub code review and Slack integration.
Solo developer running Codex daily on larger projects: Pro 5x ($100/month) — 5x Plus Codex usage plus Codex-Spark research preview; the most likely next stop after Plus burns out.
Power user who consistently hits Pro 5x limits and wants unlimited voice: Pro 20x ($200/month) — 20x Plus usage and unlimited ChatGPT Voice on desktop; only if you really need it.
Team needing SSO, RBAC, audit logs, and Slack/Linear/GitHub cloud workflows: Business / Enterprise — Sole tier exposing SAML SSO, SCIM, RBAC, compliance API, audit logs, Codex Security, and cloud integrations; seat pricing not published.
Engineer wiring Codex into CI pipelines or scripts: API Key — Pay per token via API pricing; no cloud features, which is fine for automation.

Frequently asked questions

How much does Sites cost?: Sites is included with eligible ChatGPT plans during public beta; availability depends on plan, region, and workspace settings.
What are the usage limits for my plan?: Limits depend on the model, task size and complexity, and whether work runs locally or in the cloud. Published Plus local messages per 5h: GPT-5.6 Sol 10-100, Terra 25-200, Luna 250-2,000, GPT-5.5 15-80, GPT-5.4 20-100, GPT-5.4 mini 60-350. Pro 5x offers 5x Plus usage; Pro 20x offers 20x. Enterprise/Edu with flexible pricing has no fixed limits; legacy plans share Plus per-seat limits.
What happens when you hit usage limits?: Active turns can finish subject to fair use. Plus and Pro users can buy additional credits to continue without upgrading. Business/Edu/Enterprise with flexible pricing can buy workspace credits. All users can run extra local chats via API key at standard API rates.
How does image generation count toward usage limits?: Image generation counts toward the same general usage limits and uses them ~3-5x faster than similar non-image turns; after limits it draws from credits. Not available on Free; API key users pay API pricing instead.
Where can I see my current usage limits?: Check the usage dashboard in ChatGPT, or use /status inside an active Codex CLI session. The page recommends checking every week or two.
What are tokens and credits?: Tokens measure what ChatGPT reads and writes; credits translate that into a simpler unit. After included limits, available credits let you keep working. Credit cost varies by model, context, reasoning, and tools.
What counts as Code Review usage?: Only reviews Codex runs through GitHub (e.g., @Codex tagged on a PR or automatic repo reviews). Reviews run locally or outside GitHub count toward general usage limits.
Can I extend my usage limits?: The page lists tips: shrink prompts, limit source material, match output to need, reduce AGENTS.md size, limit MCP servers, and switch to a smaller model (GPT-5.6 Terra or Luna) for routine tasks.

What we read

Every figure on this page was extracted from the vendor's own pricing page at the moment below. The text is reproduced so you can check our reading against what they publish today.

Source captured 2026-08-21 08:06 UTC

For the complete documentation index, see llms.txt. Markdown versions of documentation pages are available by appending .md to the page URL. ChatGPT Home Docs Use cases Resources Start searching Try ChatGPT Overview Features Configuration Developers Security Administration Home Get started Quickstart Use ChatGPT Get started with Work Import from another agent Foundations Prompting Personalize ChatGPT Skills & Plugins Permissions Explore What's new Models Pricing Glossary Available on ChatGPT desktop app Remote ChatGPT on the web Codex CLI Codex IDE extension Codex cloud Releases Changelog Feature Maturity Open Source Pricing options Invite friends and coworkers Frequently asked questions Feature availability Copy Page Pricing ChatGPT Work and Codex are included in your ChatGPT Free, Go, Plus, Pro, Business, Edu, or Enterprise plan ChatGPT Work and Codex share usage. ChatGPT Work usage inside ChatGPT uses the same pricing, credits, and usage limits as Codex. Pricing options Individual Business / Enterprise Free Explore Codex capabilities on quick coding tasks. $0 /month Get Free Go Use Codex for lightweight coding tasks. $8 /month Get Go Plus Power a few focused coding sessions each week. $20 /month Get Plus Codex on the web, in the CLI, in the IDE extension, and on iOS Cloud-based integrations like automatic code review and Slack integration The GPT-5.6 model family, including Sol, Terra, and Luna GPT-5.6 Luna for higher usage limits on lighter-weight or high-volume workloads Flexibly extend usage with ChatGPT credits Other ChatGPT features as part of the Plus plan Pro Choose 5x or 20x higher rate limits than Plus. From $100 /month Get Pro Everything in Plus and: Access to GPT-5.3-Codex-Spark (research preview), a fast Codex model for day-to-day coding tasks 5x or 20x more Codex usage than Plus* Unlimited ChatGPT Voice on the $200/month tier; tasks still draw from your Codex usage budget Other ChatGPT features as part of the Pro plan *Learn more about limits on both tiers. API Key Great for automation in shared environments like CI. Learn more Codex in the CLI, SDK, or IDE extension No cloud-based features (GitHub code review, Slack, etc.) Model availability follows the API models available to your key Pay only for the tokens Codex uses, based on API pricing Invite friends and coworkers Eligible users can send Codex invitations from the profile menu in the lower-left corner of the app. Choose Invite a friend on an eligible personal plan or Invite a coworker in an eligible Business workspace, enter the recipient’s email address, and send the invitation. The invitation dialog shows the current reward, recipient requirements, invite limits, and when rewards expire for your plan or promotion. Personal and Business referral programs have separate rewards and eligibility rules. Referrals aren’t currently available for ChatGPT Enterprise. From June 11 through June 24, 2026, eligible Plus and Pro users can invite up to three friends. When an eligible recipient sends their first Codex message, both people receive a banked rate-limit reset. Banked rate-limit resets are usable for 30 days after they’re granted. Business referrals use separate shared-workspace credit rewards; review the current terms before you send an invitation. Frequently asked questions How much does Sites cost? Sites is included with eligible ChatGPT plans during public beta. Availability depends on your plan, region, and workspace settings. What are the usage limits for my plan? The number of messages you can send depends on the model used, size and complexity of your tasks, and whether you run them locally or in the cloud. Small scripts or routine functions may consume only a fraction of your allowance, while larger projects, long-running tasks, or extended sessions that require the agent to hold more context will use significantly more per message. Tasks that look similar can consume different amounts of your allowance. Model choice, context, reasoning, tool use,

Pricing change log

What moved since we started tracking this page, newest first. A price that changes silently is the failure mode of every other pricing guide.

2026-08-21First published, extracted from https://developers.openai.com/codex/pricing.