Which plans include Codex?
OpenAI says Codex is included across ChatGPT plans, including Free and Go. Usage limits vary by plan.
OpenAI Help: Using Codex with your ChatGPT planPublic usage-policy ledger
Codex usage limits without a made-up timer
- Checked
- 2026-10-16
- Public rules
- 4
- Primary sources
- 2
Public Codex usage-limit mechanics, source changes, and the boundary between a provider policy and an account-specific timer. It records what OpenAI publishes and where the account-specific answer begins.
Your remaining Codex usage is account-specific
This page cannot read your allowance or reset time. Open Codex Settings or the limit banner in your signed-in client for the options available to your plan.
The public rules worth checking
What shares the same pool?
ChatGPT Work and Codex share the same pricing, credits, and usage limits. Other agentic features join that pool once their pricing takes effect; OpenAI names ChatGPT for Excel on Plus and Pro as the current case.
ChatGPT Learn: Codex pricing and usage limitsWhat changes how quickly the pool is used?
The model, the size and complexity of the task, and whether it runs locally or in the cloud all affect consumption. OpenAI says model choice, context, reasoning, tool use, retrieval, and caching change usage, so prompt length alone is not a reliable estimate.
ChatGPT Learn: Codex pricing and usage limitsWhat happens at a limit?
An active turn can finish, subject to fair-use limits. After that, Plus and Pro users can buy credits instead of upgrading; Business, Edu, and Enterprise workspaces on flexible pricing can buy workspace credits; and any user can run extra local chats with an API key at standard API rates.
ChatGPT Learn: Codex pricing and usage limitsChatGPT plans, prices and what the same money buys on the API
Prices and included usage are copied from OpenAI's pages with the date we read them. The provider publishes multiples and message estimates, not a token or dollar cap, so the rows keep the provider's wording and no cap is invented.
| Plan | Price | Included usage | Extra usage | Source |
|---|---|---|---|---|
| Free | $0/month | Limited Codex and ChatGPT Work access on desktop; the page publishes local-message estimates for Plus and Standard Business only, so the Free allowance is not itemized. | No | ChatGPT: Codex pricing and plans Read 2026-10-16 |
| Go | $8/month | More messages with tools, more uploads, more image creation, more voice chats and longer memory than Free; limited Codex access, allowance not itemized on the page. | Not stated | ChatGPT: Codex pricing and plans Read 2026-10-16 |
| Plus | $20/month | Estimated local Codex messages per five-hour window: 5 to 45 with GPT-6 Astra, 15 to 160 with GPT-6.1 Sol, 15 to 150 with GPT-6 Sol, 350 to 3,000 with GPT-6 Luna; weekly limits may also apply. | Yes, at additional credits purchased in the app; the page does not state the rate | ChatGPT: Codex pricing and plans Read 2026-10-16 |
| Pro 100 | $100/month | The lowest Pro allowance; OpenAI publishes no multiple of Plus. The Codex pricing page says Pro plans currently have no five-hour limit and gives no Pro message estimates. Ultrafast is not included. | Yes, at credits purchased in Settings, Usage; the help article does not state the dollar price per credit | ChatGPT: Codex pricing and plans Read 2026-10-16 |
| Pro 200 | $200/month | More usage than Pro 100; OpenAI publishes no multiple of Plus. New Pro 200 subscriptions not eligible for grandfathering get a lower allowance than previously offered. Pro plans currently have no five-hour limit. Ultrafast is not included. | Yes, at credits purchased in Settings, Usage; the help article does not state the dollar price per credit | ChatGPT: Codex pricing and plans Read 2026-10-16 |
| Pro 500 | $500/month | The highest included usage of the three Pro plans. Ultrafast for GPT-6 Astra and GPT-6.1 Sol uses included usage at 8x the Standard rate, then credits at 6x. Pro plans currently have no five-hour limit. | Yes, at credits purchased in Settings, Usage; Ultrafast draws credits at 6x the Standard credit rate | ChatGPT: Codex pricing and plans Read 2026-10-16 |
| Business | $25 per seat/month ($20 per seat/month billed annually) | Codex estimates for a Standard Business seat match Plus: 5 to 45 with GPT-6 Astra, 15 to 160 with GPT-6.1 Sol, 15 to 150 with GPT-6 Sol, 350 to 3,000 with GPT-6 Luna per five-hour window; flexible-pricing workspaces can buy additional workspace credits. | Yes, at additional workspace credits on flexible pricing; the page does not state the rate | ChatGPT: Codex pricing and plans Read 2026-10-16 |
What the Plus price buys at API list rates: $20 is 20 million input tokens or 4 million output tokens of GPT-6.1 Terra ($1 and $5 per million, registry updated November 9, 2026). Arithmetic on published rates, not a statement of the plan's allowance; past the plan's own limit, extra usage is what the provider's page says it is.
Plan price beside the API list cost of a workload
Pick a plan, a Claude model and a workload. The API side is list rates times the tokens you enter, the same arithmetic as the pricing calculator. The plan side is its published price; its cap is not published.
- Go price
- $8.00/month
- Included usage: More messages with tools, more uploads, more image creation, more voice chats and longer memory than Free; limited Codex access, allowance not itemized on the page.
- Same tokens on the API, list rates
- $6.00/month
- 2,000,000 input and 800,000 output tokens a month on GPT-6.1 Terra.
- Does the plan cover it?
- Unavailable
- The plan does not publish its included usage in credits or dollars; no equivalent can be computed. At this workload the API bill is below the plan price; whether the plan would also cover it depends on the unpublished cap.
Estimate only, at list rates from the pricing registry. It excludes reasoning tokens billed as output, batch discounts, taxes and provider minimums. Confirm the exact model identifier in the model ID directory and the rate on the Claude API pricing hub.
A published reset rule is not your reset clock
Product limits can depend on plan, location, workspace controls, prior use, feature availability, and the exact work requested. A useful public page preserves that distinction rather than turning a provider article into a fake personal dashboard.
Radar watches
- Published changes to plans, usage mechanics, reset rules, and model availability.
- Public Codex release and documentation changes with the original source attached.
Radar does not read
- Your remaining allowance, private reset timer, threads, prompts, or account bill.
- A promise that a public policy change changes every account at the same time.