#usage

1 messages · Page 1 of 1 (latest)

burnt kayakBOT
#

Hi @covert raft. I'm kapa.ai, your friendly support bot. I will try to answer your question by searching through the knowledge base. :wave:

#

Yes, Codex usage and ChatGPT usage are tracked separately on the Plus plan. Here's how they differ:

ChatGPT Usage (Plus)

  • Standard ChatGPT message limits apply to conversations, image generation, voice, file uploads, etc.
  • These have their own reset periods and banners (e.g., image limits, voice caps).

Codex Usage (Plus)

  • Codex has its own separate usage limits, measured within a 5-hour rolling window.
  • Limits vary by model, for example per 5-hour window on Plus: [Codex pricing limits]
Model Local Messages / 5h Cloud Tasks / 5h Code Reviews / 5h
GPT-5.5 15–80 Not available Not available
GPT-5.4 20–100 Not available Not available
GPT-5.4-mini 60–350 Not available Not available
GPT-5.3-Codex 30–150 10–60 20–50
  • The actual number of messages you can send varies based on task size and complexity — larger codebases or longer sessions consume more of your allowance per message. [Codex usage limits]

Key Point on Separation

ChatGPT file upload limits, image/video generation limits, and voice caps do not apply to Codex — they are entirely separate. [Codex FAQ]

#

Extending Usage

Once you hit either limit, you can purchase credits as a pay-as-you-go add-on to extend usage beyond your plan's included limits for both Codex and Sora. [Credits overview]

Note: you can ask me a follow up question by @ mentioning me again :speech_balloon:

#

@covert raft Kapa.ai is still learning and improving, please let me know how I did by reacting below.

minor hatch
#

@burnt kayak does the code review agent on gh use regular subscription usage bucket? i remember it used to be a distinct usage bar in codex

burnt kayakBOT
#

Let me check the knowledge base for you again. One moment! :books:

#

Code Review usage on GitHub does use the Codex usage bucket, not a separate one from your general ChatGPT subscription — but it has its own distinct sub-limit within Codex.

Specifically:

  • Code Reviews triggered via GitHub (e.g., tagging @Codex on a PR or enabling automatic reviews) count against the Code Reviews / 5h column in your Codex limits. [Codex FAQ]
  • Reviews run locally or outside of GitHub count toward your general Codex usage limits instead. [Codex FAQ]

On the Plus plan, the Code Review limit is 20–50 reviews per 5-hour window (only available on GPT-5.3-Codex). [Codex pricing limits]

So your memory is correct — Code Review does have its own distinct tracking within Codex (separate from Local Messages and Cloud Tasks), but it's still part of the overall Codex usage system, not your general ChatGPT subscription bucket.

Note: you can ask me a follow up question by @ mentioning me again :speech_balloon:

#

@minor hatch Kapa.ai is still learning and improving, please let me know how I did by reacting below.

minor hatch
#

There's no longer a usage limit signifier for Code Reviews, where can I find it?

burnt kayakBOT
#

Let me check the knowledge base for you again. One moment! :books:

#

Based on the knowledge sources available, you can check your current Codex usage limits — including Code Reviews — in the Codex usage dashboard. [Codex usage dashboard]

Additionally, if you're in an active Codex CLI session, you can use the /status command to see your remaining limits. [Codex usage dashboard]

The knowledge sources don't contain specific information about why the Code Review usage indicator may no longer appear at the analytics URL you mentioned (/codex/cloud/settings/analytics#usage). If the dashboard at chatgpt.com/codex/settings/usage doesn't show it either, it may be worth reaching out to OpenAI support directly, as this could be a UI change or a display issue similar to the credit tracking bug reported by other users. [Forum post]

Note: you can ask me a follow up question by @ mentioning me again :speech_balloon: