Codex usage: check it, know your limits, read the cost

In the Codex CLI, type /status to see your remaining limits during a session. Outside a session, open the Codex usage dashboard for your limits and reset times.

Checked against OpenAI docs on November 12, 2026.

Check your Codex usage

  • /status in the Codex CLI: your remaining limits during an active session, with the session’s configuration and token usage.
  • Usage dashboard (chatgpt.com/codex/settings/usage): your current limits and reset times.
  • API key users: the OpenAI Platform usage page shows API spend. It doesn’t show ChatGPT plan limits.

Codex limits by plan

Codex is included in ChatGPT plans. Local messages and cloud chats share your plan’s usage allowance, and cloud tasks may use more of it than local messages.

Plus and Standard Business have a five-hour window. OpenAI’s estimates of local messages per five hours:

ModelLocal messages per 5 hours (estimate)
GPT-6 Astra5 to 45
GPT-6.1 Sol15 to 160
GPT-6.1 Terra35 to 285
GPT-6 Sol15 to 150
GPT-6 Luna350 to 3,000

OpenAI says these “are not fixed message limits” and that weekly limits may also apply.

  • Pro: currently no five-hour limit. Weekly limits may apply.
  • Enterprise and Edu with flexible pricing: no fixed rate limits; usage scales with credits. Without flexible pricing, they have the same per-seat limits as Plus for most features.
  • Speed modes use more: Fast uses included usage 2.5x as fast. Ultrafast, on GPT-6 Astra and GPT-6.1 Sol, uses it 8x as fast.
  • When you hit the limit: Codex can finish the turn it’s on. Plus and Pro users can buy credits, and anyone can run extra local chats with an API key at standard API rates.

What Codex usage costs

On a plan, usage is included. Past that, credits pay for it at these rates per million tokens (Standard speed):

ModelInputCached inputOutput
GPT-6 Astra250 credits25 credits1,250 credits
GPT-6.1 Sol50 credits2.5 credits250 credits
GPT-6.1 Terra25 credits1.25 credits125 credits
GPT-6 Sol50 credits5 credits250 credits
GPT-6 Luna2.5 credits0.25 credits12.5 credits

With an API key, standard OpenAI API prices apply instead.

Why your Codex usage runs out faster than expected

  • Model and speed. Bigger models use more of the allowance per message, and Fast and Ultrafast multiply it.
  • Fresh input is the expensive part. In OpenAI’s credit rates, cached input costs a tenth of fresh input on most Codex models and a twentieth on GPT-6.1 Sol and GPT-6.1 Terra.
  • Context on every message. A long AGENTS.md, extra files in context and MCP servers you aren’t using all add tokens to each request.

Track Codex usage across sessions

Codex keeps session logs in ~/.codex/sessions (and ~/.codex/archived_sessions). npx sigrank reads them and shows input, output and cache counts across the available time windows, and sigrank compare --platform codex checks those counts across your local sources. Codex logs count cache writes inside input, so SigRank estimates the split for Codex; the CLI reference documents how.

Usage tells you how much. Yield tells you how well.

Same limit, very different output. SigRank scores it with Yield: cache read × output ÷ input². On GPT-6.1 Sol a cached token costs a twentieth of a fresh one, so reusing context is where the score comes from. How Yield works

Run npx sigrank to see yours. Nothing is published until you run sigrank submit, and sigrank submit --dry-run shows the exact payload first. Prompts, responses, file paths and repo or project names are never sent.

npx sigrankSee the board →

FAQ

How do I check my Codex usage?
Type /status in the Codex CLI during a session, or open the Codex usage dashboard at chatgpt.com/codex/settings/usage for limits and reset times.
When does Codex usage reset?
On Plus and Standard Business, the five-hour window resets five hours after it starts. Weekly limits, where they apply, reset on their own schedule. The usage dashboard shows the reset times.
Is Codex usage separate from ChatGPT?
No. Codex is included in your ChatGPT plan, and local messages and cloud chats share the plan's usage allowance.
How much Codex usage comes with ChatGPT Plus?
OpenAI estimates 15 to 160 local messages per five hours on GPT-6.1 Sol, depending on the task. It is not a fixed number.
Does Codex Pro have a 5-hour limit?
Not currently. Weekly limits may apply.
What happens when I hit the Codex limit?
Codex can finish the turn it is on. Then you can wait for the reset, buy credits on Plus or Pro, or run local chats with an API key at standard API rates.
Does SigRank read my code?
No. Prompts, responses, file paths and repo or project names are never sent, and nothing is sent until you run sigrank submit.