Quick answer
Codex documents pro 5× local-message range as 50–10,000 messages per 5 hours, depending on model. The range spans GPT-5.6 Sol through the lighter GPT-5.6 Luna in the official plan table.
Verified Aug 22, 2026Official source
Current limits
| Constraint | Current value | Scope | Verified source |
|---|---|---|---|
| Pro 5× local-message rangeThe range spans GPT-5.6 Sol through the lighter GPT-5.6 Luna in the official plan table. | 50–10,000 messages per 5 hours, depending on model | ChatGPT Pro 5× | OpenAIAug 22, 2026 |
Why does this limit matter?
A smaller model can extend sustained local work when the highest-capability model is unnecessary.
This value is scoped to ChatGPT Pro 5×; a different plan, runtime, model, endpoint, region, or account can produce a different effective constraint.
What should you check?
- Match the selected model to the current pricing table and verify remaining usage with /status.
- Confirm the exact plan, model, runtime, endpoint, region, and account that serve the failing workload.
- Record the observed value, response headers or configuration, timestamp, and source without logging secrets.
Important caveats
- Task size, context, reasoning, tools, images, and speed settings change actual consumption.
- Treat the official source and live account configuration as authoritative if they differ from this verified snapshot.
HyperObserve reports the documented platform constraint. Your application, SDK, gateway, provider, region, or account can impose a lower effective limit.
Related