If you are looking up Codex pricing, you need the monthly cost, which ChatGPT plan includes OpenAI’s coding agent, and how usage limits work. Codex is the agent that can write and edit code in ChatGPT, in the Codex CLI (command-line interface), in an IDE extension, and on related cloud surfaces. ChatGPT Work is the agentic work surface inside ChatGPT.
OpenAI’s Codex pricing page states that ChatGPT Work and Codex are included in ChatGPT Free, Go, Plus, Pro, Business, Edu, or Enterprise. Work usage inside ChatGPT uses the same pricing, credits, and usage limits as Codex. You pay on a ChatGPT plan, including Free, or with an API key. An API key runs Codex in the CLI, SDK, or IDE extension, including automation such as CI, billed at API pricing. Cloud features such as GitHub code review and Slack are excluded on that path.
Plus is $20 a month. Pro starts at $100 a month with a 5x or 20x usage choice; the pricing page lists no separate monthly dollar prices for those two tiers, and the usage difference is the five-hour estimate table. Business is $20 per user per month billed annually for two or more users, or $25 per user per month billed monthly. The five-hour table’s Standard Business column matches Plus. A separate note says “Business ($100) uses the Pro 5x estimates”; the page does not define that $100 label or connect it to the $20/$25 card. Free and Go are $0 and $8, with GPT-6 Luna at Standard speed in the desktop app, subject to rollout, and they have no column in the five-hour table.
Two dates on that page matter for planning. GPT-5.5 retires from ChatGPT, ChatGPT Work, and Codex on all plans on October 14, 2026, and OpenAI states the OpenAI API isn’t affected. GPT-5.6 Sol promotional pricing lasts at least through November 21, 2026. The five-hour message ranges in this guide are official estimates; current limits and reset times are on the usage dashboard.
Codex is billed as a ChatGPT-plan feature
Codex is OpenAI’s coding agent. It can run in ChatGPT, in the Codex CLI (the command-line interface), in an IDE extension, on iOS on some plans, and through cloud integrations such as automatic code review and Slack. ChatGPT Work is the related agent surface inside ChatGPT. OpenAI bills them as one product family on a ChatGPT plan, from Free through Enterprise, or on an API key.
A ChatGPT plan is how you get Codex as an included feature, with plan-specific surfaces and, on some plans, optional credits after the included allowance. An API key is how you run Codex for automation in shared environments such as CI (continuous integration). On the API-key path, Codex stays in the CLI, SDK, or IDE extension, the models you can call follow the API models on that key, and usage is billed at API pricing. Cloud features such as GitHub code review and Slack are excluded on that path.
Plan prices, Codex access, and limit style at a glance
OpenAI’s Codex pricing page, as of 2026-09-27, lists these plan prices and access notes. Five-hour message figures later in this article are official estimates. Live remaining quota is on the usage dashboard.
Codex access and list prices from OpenAI’s Codex pricing page (as of 2026-09-27). Five-hour message figures later in this article are official estimates; live remaining quota is on the usage dashboard.
| Plan | Official price | Codex surfaces on the pricing card | How usage is billed |
| — | — | — | — |
| Free | $0/month | GPT-6 Luna at Standard speed in the desktop app, subject to rollout | No column in the five-hour estimate table |
| Go | $8/month | GPT-6 Luna at Standard speed in the desktop app, subject to rollout | No column in the five-hour estimate table |
| Plus | $20/month | Web, CLI, IDE extension, and iOS; cloud integrations such as automatic code review and Slack; GPT-6 Sol and GPT-6 Luna | Included limits; optional ChatGPT credits |
| Pro | From $100/month | Everything in Plus; choose 5x or 20x higher rate limits than Plus | 5x or 20x Codex usage versus Plus; optional credits. The 5x and 20x choices appear without their own monthly dollar prices on this page |
| Business | $20 per user per month billed annually (2+ users), or $25 per user per month billed monthly | ChatGPT and Codex on desktop and mobile; larger virtual machines for cloud chats | Standard Business five-hour ranges match Plus; optional credits |
| Enterprise and Edu | Contact sales | Everything in Business, plus priority processing and enterprise admin, security, and compliance controls | Flexible pricing: usage scales with credits. Without flexible pricing: same per-seat limits as Plus for most features |
| API key | Usage charged at API pricing | CLI, SDK, or IDE extension only; no cloud features such as GitHub code review and Slack | API pricing; marked usage-based in the estimate table |
Free, Go, Plus, Pro, and the API-key path
The Free pricing card is $0 per month and describes exploring Codex on quick coding tasks, with GPT-6 Luna at Standard speed in the desktop app, subject to rollout. The Go card is $8 per month and describes lightweight coding tasks, with the same Luna desktop rollout note. The five-hour estimate table publishes no Free or Go column.
The Plus card is $20 per month and describes powering a few focused coding sessions each week. It lists Codex on the web, in the CLI, in the IDE extension, and on iOS; cloud-based integrations such as automatic code review and Slack; GPT-6 Sol and GPT-6 Luna; and the option to extend usage with ChatGPT credits.
Pro starts at $100 per month. It includes everything in Plus, and you choose 5x or 20x higher rate limits than Plus, described as 5x or 20x more Codex usage than Plus. The pricing page lists no separate monthly dollar prices for the 5x and 20x choices. The usage difference is the five-hour estimate table; use that table for those two tiers rather than multiplying a Plus message count yourself.
An API key is aimed at automation in shared environments such as CI. You get Codex in the CLI, SDK, or IDE extension. Cloud-based GitHub code review and Slack are excluded. Model availability follows the API models on the key, and you pay for usage based on API pricing. The Codex pricing page does not list USD API token rates.
Business, Enterprise, and Edu
Business is $20 per user per month when billed annually for two or more users, or $25 per user per month billed monthly. The Business card lists ChatGPT and Codex across desktop and mobile apps, larger virtual machines to run cloud chats faster, credits to extend usage, a dedicated workspace with essential admin controls, SAML SSO (single sign-on), and MFA (multi-factor authentication), and no training on business data by default.
The five-hour table’s Standard Business column matches the Plus ranges. Under that table, OpenAI writes: “Business ($100) uses the Pro 5x estimates.” The page does not define that $100 label or connect it to the $20/$25 Business card.
Enterprise and Edu are contact-sales plans. They include everything in Business plus priority request processing; SCIM, EKM, user analytics, domain verification, and role-based access control (RBAC); audit logs and usage monitoring via the Compliance API; and data retention and data residency controls. Enterprise and Edu users with flexible pricing have no fixed rate limits; usage scales with credits. Enterprise and Edu plans without flexible pricing have the same per-seat usage limits as Plus for most features.
Usage limits are estimates, with five-hour local-message ranges
How many messages you can send depends on the model, the size and complexity of the task, and whether you run it locally or in the cloud. Small scripts may use a fraction of the allowance. Larger projects, long-running tasks, or sessions that hold more context use more per message. Similar-looking tasks can consume different amounts because model choice, context, reasoning, tool use, retrieval, and caching all affect usage. Prompt length alone is not a reliable estimate.
The published table shows local messages per five-hour period. OpenAI labels those figures as estimates and sends you to the usage dashboard for current limits and reset times. Cloud chats on ChatGPT plans use GPT-5.6 Sol and may use more of the allowance than local messages. Local messages and cloud chats share the plan’s usage allowance. Weekly limits may also apply. Usage limits are shared with other agentic features once those features are priced; the page currently names ChatGPT for Excel on Plus and Pro.
Cloud chats on ChatGPT plans may use more of the allowance than local messages. Fast mode uses 2.5x the Standard credit rate for GPT-6 Astra, Sol, and Luna where available. Image generations use included limits about 3–5x faster on average than similar turns without image generation, depending on quality and size. Desktop Voice shares the same allowance at a per-minute rate. GPT-5.5 still appears in this estimate table; it retires from ChatGPT, ChatGPT Work, and Codex on all plans on October 14, 2026.
Official estimates of local messages per five-hour period, from OpenAI’s Codex pricing page (as of 2026-09-27). OpenAI labels these as estimates. Current limits and reset times are on the usage dashboard. Standard Business matches Plus. The table has no Free or Go column.
| Model | Plus | Pro 5x | Pro 20x | Standard Business | API Key |
| — | — | — | — | — | — |
| GPT-6 Astra | 5-45 | 25-225 | 100-900 | 5-45 | Usage-based |
| GPT-6 Sol | 15-150 | 70-700 | 300-3,000 | 15-150 | Usage-based |
| GPT-6 Luna | 350-3,000 | 1,750-14,000 | 7,000-56,000 | 350-3,000 | Usage-based |
| GPT-5.6 Sol | 10-100 | 50-500 | 200-2,000 | 10-100 | Usage-based |
| GPT-5.6 Terra | 25-200 | 125-1,000 | 500-4,000 | 25-200 | Usage-based |
| GPT-5.6 Luna | 250-2,000 | 1,250-10,000 | 5,000-40,000 | 250-2,000 | Usage-based |
| GPT-5.5 | 15-80 | 75-400 | 300-1,600 | 15-80 | Usage-based |
| GPT-5.4 | 20-100 | 100-500 | 400-2,000 | 20-100 | Usage-based |
| GPT-5.4 mini | 60-350 | 300-1,750 | 1,200-7,000 | 60-350 | Usage-based |
Sites, Voice, images, Fast mode, and what happens after you hit the limit
Sites is included with eligible ChatGPT plans during public beta. Availability depends on plan, region, and workspace settings.
Desktop Voice uses the existing Codex usage budget at $0.05 per minute. GPT-Live manages the live conversation. The model handling the task is billed separately at its standard token rates. Voice and tasks share the plan’s usage limits. For Business, Edu, and Enterprise workspaces with credit-based billing, desktop voice costs 1.25 credits per minute. The same credit rate applies when Plus and Pro users spend additional credits. ChatGPT Voice in Desktop is unavailable via API key.
Image generation counts toward the same general usage limits as local messages and cloud chats. OpenAI says image generations use included limits 3–5x faster on average than similar turns without image generation, depending on quality and size. After included limits, image generation draws from credits. Image generation is unavailable on Free. When Codex runs on an API key, API pricing applies to image generation.
Speed configurations increase credit consumption for applicable models, so they also use included limits faster. Fast mode uses 2.5x the Standard credit rate for GPT-6 Astra, Sol, and Luna where available. OpenAI’s Speed documentation lists supported models and rates.
If you reach usage limits during an active turn, the agent can continue that turn, subject to fair use. Plus and Pro users can purchase additional credits without upgrading. Business, Edu, and Enterprise plans with flexible pricing can purchase additional workspace credits. All users may run extra local chats using an API key, with usage charged at standard API rates. Current limits are in the usage dashboard. OpenAI advises checking that dashboard every week or two.
/statusTokens, credits, and Standard-speed credit rates
Tokens are the units ChatGPT reads and writes. Prompts, files, chat history, tool results, and responses all use tokens. Credits pay for eligible usage on credit-based plans after included limits. Credit purchase prices and discounts depend on the plan or agreement; the Codex pricing page does not list credit-pack dollar prices.
The rates below are for Standard speed, in credits per million input tokens, cached input tokens, and output tokens. Codex credit billing has no separate cache-write charge. API-key usage follows API pricing, not this credit table.
GPT-5.6 usage averages 5–30 credits per message. Credit prices alone do not determine included subscription usage. GPT-5.6 Sol promotional pricing is available at least through November 21, 2026. OpenAI also states that GPT-5.6 Sol, Terra, and Luna rates remain unchanged.
Daybreak access requires Trusted Access for Cyber approval. Daybreak Blue uses GPT-5.6 Sol credit rates. Daybreak Red requires separate approval and provisioning. GPT-Rosalind-Research appears on the rate card without a further definition on this page. A small subset of Enterprise customers should continue using the legacy rate card until migrated to token-based pricing. If an Enterprise agreement specifies usage-based billing in USD, that agreement and the Enterprise USD rate card apply instead of the credit table.
Standard-speed Codex credit rates per million tokens, from OpenAI’s Codex pricing page (as of 2026-09-27). Codex credit billing has no separate cache-write charge. API-key usage follows API pricing.
| Model | Input credits / 1M tokens | Cached input credits / 1M tokens | Output credits / 1M tokens |
| — | — | — | — |
| GPT-6 Astra | 250 | 25 | 1,250 |
| GPT-6 Sol | 50 | 5 | 250 |
| GPT-6 Luna | 2.5 | 0.25 | 12.5 |
| GPT-5.6 Sol | 100 | 10 | 500 |
| Daybreak Blue | 100 | 10 | 500 |
| Daybreak Red | 312.5 | 31.25 | 1875 |
| GPT-5.6 Terra | 50 | 5 | 300 |
| GPT-5.6 Luna | 5 | 0.5 | 30 |
| GPT-Rosalind-Research | 125 | 12.5 | 625 |
| GPT-5.5 | 125 | 12.50 | 750 |
| GPT-5.4 | 62.50 | 6.250 | 375 |
| GPT-5.4 mini | 18.75 | 1.875 | 113 |
| GPT-Image-2 (image) | 200 | 50 | 750 |
| GPT-Image-2 (text) | 125 | 31.25 | 250 |
Code Review, stretching included limits, and what this page leaves out
Code Review usage applies only when Codex runs reviews through GitHub, such as tagging @Codex on a pull request or enabling automatic repository reviews. Reviews run locally or outside GitHub, including local /review, count toward general usage limits.
The published usage limits and credits are average rates. The bullets below are OpenAI’s tips for making those limits last longer.
The Plus and Pro pricing cards list Codex on the web, CLI, IDE extension, and iOS; cloud-based integrations such as automatic code review and Slack; GPT-6 Sol and GPT-6 Luna; and optional credits. The feature table also lists Codex cloud, ChatGPT Work on the web, desktop local chats, SDK and scripting, Fast mode, image generation, Voice, local review, and cloud GitHub, Slack, and Linear integrations for those plans. Some browser and Computer Use rows are marked Limited* to specific regions. Limited† means some first-party plugins are unavailable. Geographic details live in the linked feature docs.
The API-key card excludes cloud-based GitHub code review and Slack. Sites is included with eligible ChatGPT plans during public beta; availability depends on plan, region, and workspace settings. Business adds workspace admin, SAML SSO, and MFA. Enterprise and Education add SCIM, EKM, RBAC, analytics, the Compliance API, and retention and residency controls.
The pricing page does not publish a live remaining-quota number for a given account, USD API list prices for API-key token usage, credit-pack purchase prices, or a ranking against other coding tools. Remaining quota belongs on the usage dashboard. Feature-specific Speed, geography, and GPT-5.5 migration details live on those official docs.
- Keep prompts precise and drop extra context.
- Attach only relevant files, and narrow sources or date range when you can.
- Define audience, format, and length, and separate required work from optional polish.
- Nest smaller AGENTS.md files (the repo instruction files the agent reads) so a large project does not inject one oversized file.
- Run fewer MCP servers (Model Context Protocol servers that plug extra tools into the agent), and disable them when you do not need them, because each server adds context.
Open OpenAI’s Codex pricing page and match your ChatGPT plan to the access card. Then check the usage dashboard for remaining quota and reset times. In an active Codex CLI session, run /status. If you are on Plus or Pro and hit the included limit, buy additional credits or continue local chats with an API key at API rates. Flexible Business, Edu, and Enterprise plans can buy workspace credits. For Pro, read the five-hour estimate table for 5x versus 20x; the page lists no separate monthly prices for those choices.