GPT-6 Astra Usage Limits Explained: Plus vs Pro, Weekly Quotas, Resets & Message Limits
GPT-6 Astra Usage Limits: Plus vs Pro, Resets & Quotas
If you have started using GPT-6 Astra and noticed your available usage disappearing faster than expected, you are not necessarily seeing a bug.
OpenAI applies different GPT-6 Astra usage limits depending on your ChatGPT plan, the product you are using, the size of your task and the reasoning settings you select.
The most important thing to understand is that there are actually two different limit systems to keep separate:
- GPT-6 Astra in ChatGPT Work and Codex, which can be governed by five-hour and weekly usage allowances.
- GPT-6 Pro in regular Chat, which is powered by Astra but has its own message limits.
For example, ChatGPT Plus users can access Astra through Work and Codex, but Plus does not currently include GPT-6 Pro in standard Chat. Pro and eligible Business or Enterprise users can access GPT-6 Pro in Chat.
This guide explains the current GPT-6 Astra limits for Plus, Pro $100, Pro $200, Business and API users, when the limits reset, why Astra can consume usage faster than other models, and what you can do when you reach your quota.
The information below is based primarily on OpenAI’s current GPT-6 Astra usage documentation and GPT-6 Pro ChatGPT limits.
Last updated: September 14, 2026.
GPT-6 Astra Usage Limits: Quick Answer
OpenAI currently estimates the following number of local messages per five-hour period for GPT-6 Astra in Work and Codex:
| Plan | Estimated GPT-6 Astra Local Messages Per 5 Hours |
|---|---|
| ChatGPT Plus | 5–45 |
| Pro 5X / Pro $100 | 25–225 |
| Pro 20X / Pro $200 | 100–900 |
| Business Standard | 5–45 |
| API | Usage-based |
These are not guaranteed fixed message limits.
According to OpenAI, actual usage varies depending on the complexity of the task, input and output length, reasoning level, Fast mode and the amount of work performed.
A large multi-step coding or research task can therefore consume considerably more allowance than a short request.
OpenAI also applies a weekly allowance in some cases, meaning you may have enough remaining usage in your five-hour window but still encounter a weekly limit.
You can see your own current allowance and reset time by opening:
ChatGPT → Settings → Usage
Source: OpenAI — Managing usage with GPT-6 Astra
What Is GPT-6 Astra?
GPT-6 Astra is OpenAI’s frontier model designed for demanding work involving areas such as complex reasoning, software engineering, research, computer use, analysis and document creation.
OpenAI describes Astra as its most capable model for difficult end-to-end work.
For developers, the model is available through the API as:
gpt-6-astra
OpenAI’s API documentation currently lists a 1,050,000-token context window and up to 128,000 output tokens for the model.
Astra also supports capabilities including web search, file search, code execution, computer use and other tool-based workflows.
Developers can see the official specifications on the GPT-6 Astra API model page.
The arrival of models capable of performing increasingly complex AI-development tasks is also relevant to the wider debate around AI systems helping build future AI. We cover that subject separately in our guide to self-improving AI and recursive self-improvement.
GPT-6 Astra Plus Limit
ChatGPT Plus users have access to GPT-6 Astra through ChatGPT Work and Codex.
OpenAI estimates that Plus users may receive approximately:
5–45 Astra local messages per five-hour window.
However, you should not interpret this as meaning that every Plus subscriber is guaranteed 45 Astra prompts every five hours.
The actual amount depends on what you ask Astra to do.
A short request may use relatively little allowance.
A complicated task involving:
- large files,
- long outputs,
- extensive reasoning,
- multiple tool calls,
- coding agents,
- research steps,
- or Fast mode
can use substantially more.
This explains why two Plus users can perform very different numbers of tasks before reaching a limit.
OpenAI explicitly says the local-message numbers are estimates rather than fixed caps.
Does ChatGPT Plus Get GPT-6 Pro in Chat?
Currently, ChatGPT Plus includes Astra in Work and Codex but not GPT-6 Pro in ordinary Chat.
GPT-6 Pro, which is powered by GPT-6 Astra, is available in ChatGPT Chat for eligible:
- Pro $100 users,
- Pro $200 users,
- Business users,
- Enterprise users.
Availability can also depend on workspace permissions.
That distinction is important because someone may have Astra available in Codex or Work but not see GPT-6 Pro in the normal Chat model picker.
Source: OpenAI — GPT-5.6 and GPT-6 Pro in ChatGPT
GPT-6 Astra Pro $100 Limit
OpenAI’s Pro $100 plan provides 5X the usage allowance of Plus.
For GPT-6 Astra in Work and Codex, OpenAI estimates approximately:
25–225 local messages every five hours.
The Pro $100 plan also includes GPT-6 Pro in standard Chat.
However, the Chat limit is separate from the Work and Codex allowance.
For GPT-6 Pro in Chat, OpenAI currently provides Pro $100 users:
50 messages per week.
That allowance is shared with GPT-5.6 Sol Pro.
Therefore, if you use the shared weekly allowance on GPT-5.6 Sol Pro, you may have less available for GPT-6 Pro.
Switching between those two models does not create a fresh quota after the shared allowance has been exhausted.
GPT-6 Astra Pro $200 Limit
ChatGPT Pro $200 is also called Pro 20X because it offers up to 20 times the usage allowance of Plus for applicable features.
For Astra in Work and Codex, OpenAI currently estimates:
100–900 local messages per five-hour period.
For GPT-6 Pro in standard Chat, the Pro $200 plan currently receives:
200 GPT-6 Pro messages per week.
GPT-5.6 Sol Pro has a separate allowance of 170 messages per day, while OpenAI also applies a combined daily limit of 200 messages across the relevant Pro-model usage.
Once the weekly GPT-6 Pro allowance has been reached, ChatGPT can automatically switch the conversation to GPT-5.6 Thinking at Medium.
Source: OpenAI ChatGPT model and limit documentation
Is ChatGPT Pro $200 Currently Available?
There is an important temporary restriction.
As of September 10, 2026, OpenAI has temporarily paused new subscriptions and upgrades to ChatGPT Pro $200 / Pro 20X.
Existing Pro $200 subscribers are not affected and can continue using their subscriptions.
Pro $100 remains available.
According to OpenAI, users on Free, Go, Plus or Pro $100 cannot currently upgrade to Pro $200 while the pause is in effect.
OpenAI has not provided a confirmed public date for when new Pro $200 subscriptions will reopen.
You can check the latest status on OpenAI’s official ChatGPT Pro tiers page.
Because this situation can change, readers considering Pro 20X should check OpenAI’s page directly before making a subscription decision.
GPT-6 Astra Business Limits
OpenAI also makes Astra available to eligible Business users.
For a Business Standard seat, OpenAI currently estimates:
5–45 Astra local messages per five-hour period.
Business Premium users receive substantially higher included Work and Codex usage.
Business plans can also use workspace credits for eligible additional Astra usage when those credits are available and workspace spending settings permit it.
In standard Chat, GPT-6 Pro allowances are different.
OpenAI currently lists:
| Business Plan | GPT-6 Pro Chat Allowance |
|---|---|
| Business Standard | 15 messages per month |
| Business Premium | 50 messages per week |
These limits relate to GPT-6 Pro in Chat, not the separate Astra allowance used by Work and Codex.
That distinction is one of the most important details to understand when comparing ChatGPT plans.
GPT-6 Astra Five-Hour Limit Explained
The five-hour window is not simply a timer that gives everyone the same number of Astra prompts.
OpenAI describes it as a usage allowance covering a five-hour period.
A new five-hour window begins when you send your first eligible Work or Codex request after the previous window has ended.
Within that window, your usage depends on the amount of computational work being performed.
For example, someone asking Astra to review a small function may use considerably less allowance than someone asking it to:
- inspect an entire repository,
- perform web research,
- analyze several documents,
- make code changes,
- execute tools,
- test its work,
- and generate a long report.
That is why OpenAI provides a range such as 5–45 messages instead of guaranteeing one exact prompt count.
Does GPT-6 Astra Have a Weekly Limit?
Yes, a weekly limit may apply in addition to the five-hour usage window.
When both limits apply, you must have allowance remaining in both.
This creates a situation where you could technically be inside a new five-hour period but still be unable to continue because your weekly allocation has already been consumed.
OpenAI recommends checking:
Settings → Usage
to determine which limit you reached and when it will reset.
Switching to another model does not necessarily restore Astra usage if the models are drawing from the same shared allowance.
When Does the GPT-6 Astra Limit Reset?
There is no single universal reset time that applies identically to every Astra user.
For Work and Codex, OpenAI uses a combination of usage windows.
The five-hour window begins when you make your first request after the previous window has ended.
Your account may also have a weekly usage period.
OpenAI displays the relevant remaining allowance and reset information under:
Settings → Usage
This is more reliable than calculating your reset manually because usage structures may differ by plan or account.
Why Does GPT-6 Astra Use My Limit So Fast?
OpenAI explicitly warns that Astra can consume your included allowance faster than GPT-5.6 Sol.
Several factors can increase Astra usage.
1. Large prompts
Submitting large files, repositories or long conversation histories requires more processing.
2. Long outputs
Tasks that produce extensive code, analysis or documents can use more allowance.
3. Higher reasoning levels
Higher reasoning effort can require more computation.
OpenAI offers Astra reasoning settings including lower and higher effort levels. For many tasks, starting with Low or Medium can make the allowance last longer.
4. Multi-step tasks
A task that browses the web, reads files, runs code and revises its own output performs more work than a basic question-and-answer interaction.
5. Fast mode
OpenAI says Fast mode can use more of your included allowance.
For Astra specifically, OpenAI’s current rate card lists Fast mode in Work and Codex at a higher usage multiplier.
Therefore, users trying to maximize the number of tasks they can complete may not always want to use the highest reasoning level or fastest mode for routine jobs.
GPT-6 Astra vs GPT-5.6 Sol Usage Limits
One of the easiest ways to understand Astra’s cost in terms of allowance is to compare it with OpenAI’s other models.
OpenAI currently provides these approximate local-message ranges per five-hour period:
| Model | Plus | Pro 5X | Pro 20X |
|---|---|---|---|
| GPT-6 Astra | 5–45 | 25–225 | 100–900 |
| GPT-5.6 Sol | 10–100 | 50–500 | 200–2,000 |
| GPT-5.6 Terra | 25–200 | 125–1,000 | 500–4,000 |
| GPT-5.6 Luna | 250–2,000 | 1,250–10,000 | 5,000–40,000 |
The comparison shows why Astra can feel significantly more limited.
It is designed for particularly difficult work rather than maximizing the number of lightweight requests.
For routine editing, extraction, categorization or simple drafting, a less expensive model may therefore make more sense.
For difficult debugging, research, analysis or complex end-to-end tasks, Astra may justify its heavier allowance usage.
How to Make Your GPT-6 Astra Limit Last Longer
There is no legitimate trick that gives you unlimited Astra usage, but you can use your included allowance more efficiently.
Use Astra only when the task needs it
You do not necessarily need OpenAI’s most capable model to rewrite a short paragraph or extract five names from a document.
Use less expensive models for simpler work and reserve Astra for harder tasks.
Start with lower reasoning
OpenAI specifically suggests trying Astra at Low or Medium reasoning before automatically selecting the highest setting.
Higher reasoning does not guarantee a better answer for every task.
Give clear instructions
A vague prompt may force the model to spend additional steps interpreting the task or revising unsuccessful work.
A clear prompt can reduce unnecessary work.
Provide the right files immediately
If the model lacks essential information, increasing the reasoning setting will not magically give it access to missing data.
Provide the required files, context and permissions before starting.
Check usage before major tasks
Go to:
Settings → Usage
before launching a large task.
That can prevent a complex workflow from stopping midway because your allowance is nearly exhausted.
What Happens When You Reach the GPT-6 Astra Limit?
If you reach an Astra usage limit, your available options depend on your account.
OpenAI says users may be able to:
- wait for the limit to reset,
- use another available model,
- use a saved reset,
- purchase an eligible instant reset,
- or continue using eligible purchased credits.
The options actually shown depend on your subscription, billing country and account eligibility.
OpenAI also warns that simply switching models does not restore allowance if those models share the same usage pool.
What Is a GPT-6 Astra Banked Reset?
During the initial Astra rollout, OpenAI gave some eligible Plus, Pro and Business users banked resets.
A banked reset is essentially a saved refresh of eligible Work and Codex usage allowances.
A full banked reset can refresh both the five-hour and weekly usage limits and change the weekly reset date.
OpenAI says eligible users received banked resets on September 3 and September 4, 2026, while a separate global reset occurred on September 7.
The September 7 reset applied automatically and did not appear as a saved reset.
You can read OpenAI’s full explanation in How banked Codex resets work.
Can You Buy More GPT-6 Astra Usage?
Some eligible personal ChatGPT Plus and Pro accounts can purchase instant resets, depending on account and billing-country availability.
A purchased reset immediately refreshes the applicable five-hour and weekly Work/Codex allowance.
Unlike a banked reset, a purchased instant reset cannot be stored for later.
OpenAI also allows certain users or organizations to continue eligible usage with purchased credits.
Before paying for additional usage, check the options displayed in your account because availability and pricing can vary.
Where Can You Check GPT-6 Astra Usage?
OpenAI recommends checking usage directly inside ChatGPT.
Go to:
Settings → Usage
There you may see:
- remaining allowance,
- usage percentage,
- five-hour reset timing,
- weekly reset information,
- available saved resets,
- and eligible credit options.
This is the best place to check your own account because the public message ranges are estimates rather than personalized guarantees.
GPT-6 Astra Missing From ChatGPT: How to Fix It
If Astra should be available on your plan but is missing, OpenAI recommends updating your ChatGPT desktop application.
Open:
Menu → Check for Updates
Install any available update, fully quit ChatGPT and reopen it.
OpenAI specifically recommends checking for updates again even if you recently updated the app.
If you use the Codex CLI, GPT-6 Astra currently requires Codex CLI version 0.153.0 or newer.
Access can still depend on your subscription and workspace permissions.
If updating does not solve the issue, verify that you are signed into the correct account and workspace.
GPT-6 Pro Chat Limits vs Astra Work Limits
This is worth repeating because it causes considerable confusion.
GPT-6 Pro in Chat and GPT-6 Astra in Work/Codex do not use exactly the same limit system.
For Chat:
| Plan | GPT-6 Pro Chat Limit |
|---|---|
| Pro $100 | 50 messages/week |
| Pro $200 | 200 messages/week |
| Business Standard | 15 messages/month |
| Business Premium | 50 messages/week |
For Work/Codex, the Astra allowance is based on the separate five-hour and weekly usage structure described earlier.
Therefore, reaching your Work/Codex Astra allowance does not necessarily mean you have simultaneously exhausted every ChatGPT feature.
Likewise, exhausting GPT-6 Pro Chat messages does not mean the entire account becomes unusable.
GPT-6 Astra API Rate Limits
Developers using the OpenAI API have a separate usage and billing system.
OpenAI currently lists GPT-6 Astra standard API pricing at:
$10 per 1 million input tokens
and
$50 per 1 million output tokens
The current API rate-limit table lists:
| API Tier | Requests Per Minute | Tokens Per Minute |
|---|---|---|
| Tier 1 | 500 | 500,000 |
| Tier 2 | 5,000 | 1,000,000 |
| Tier 3 | 5,000 | 2,000,000 |
| Tier 4 | 10,000 | 4,000,000 |
| Tier 5 | 15,000 | 40,000,000 |
Free API usage is not supported for GPT-6 Astra according to the current model page.
API usage is separate from the included ChatGPT subscription allowance.
If you sign into Codex with your ChatGPT account, ChatGPT plan limits apply.
If you use your own API key, API pricing and rate limits apply instead.
For current developer specifications and pricing, see the official GPT-6 Astra API documentation.
Why Astra’s Limits Matter for the Future of AI
The arrival of models that can spend substantial compute performing long, autonomous workflows represents a broader change in how AI is used.
Instead of answering only individual questions, frontier models increasingly perform extended coding, research, analysis and computer-use tasks.
That raises questions not only about compute and pricing but also about how increasingly capable AI systems are supervised.
For the broader technical debate, see our explainer on what AI alignment means.
We also cover how AI systems are increasingly participating in AI research itself in our guide to self-improving AI.
And for readers following one of OpenAI’s biggest competitors, our Anthropic IPO and Claude 5.1 guide tracks the latest developments around Anthropic.
GPT-6 Astra Usage Limits FAQ
What is the GPT-6 Astra limit on ChatGPT Plus?
OpenAI estimates approximately 5–45 local Astra messages per five-hour window for Plus users in Work and Codex. Actual usage varies significantly depending on the task.
Does ChatGPT Plus have GPT-6 Astra?
Yes. Plus includes GPT-6 Astra in ChatGPT Work and Codex. GPT-6 Pro in standard Chat is currently available to eligible Pro, Business and Enterprise users rather than Plus.
What is the GPT-6 Astra Pro $100 limit?
For Work and Codex, OpenAI estimates approximately 25–225 local Astra messages per five hours. Pro $100 users also receive GPT-6 Pro in Chat with a current allowance of 50 messages per week, shared with GPT-5.6 Sol Pro.
What is the GPT-6 Astra Pro $200 limit?
OpenAI estimates approximately 100–900 local Astra messages per five hours in Work and Codex for Pro 20X. GPT-6 Pro in Chat currently has a separate limit of 200 messages per week.
When does the GPT-6 Astra limit reset?
The five-hour Work/Codex window begins with your first request after the previous window expires. Weekly limits may also apply. Check Settings → Usage for the reset time specific to your account.
Why does Astra run out so quickly?
Astra is computationally intensive. Large inputs, long outputs, high reasoning levels, Fast mode and multi-step agent tasks can consume significantly more allowance.
Does switching models reset Astra usage?
Not necessarily. OpenAI says switching models does not restore usage when the models share the same allowance pool.
Can I buy more Astra usage?
Eligible users may be offered additional credits or purchased instant resets. Availability depends on your plan, account and billing country.
Does GPT-6 Astra have unlimited usage on Pro?
No. OpenAI explicitly applies usage allowances to Astra and GPT-6 Pro. A Pro subscription does not mean unlimited GPT-6 Pro usage.
Is Pro $200 available right now?
As of September 14, 2026, OpenAI says new Pro $200 / Pro 20X subscriptions and upgrades remain temporarily paused. Existing Pro $200 subscriptions continue normally.
Does GPT-6 Astra have a weekly limit?
A weekly allowance may apply in addition to the five-hour Work/Codex window. Your account must have enough usage remaining in all applicable windows.
Where can I see my remaining Astra usage?
Open ChatGPT → Settings → Usage to see current allowance information and reset times available for your account.
Is GPT-6 Pro the same as GPT-6 Astra?
GPT-6 Pro in Chat is powered by GPT-6 Astra, but OpenAI treats Chat usage separately from Astra usage in Work and Codex.
Is GPT-6 Astra available through the API?
Yes. Developers can access the model through the OpenAI API using gpt-6-astra, subject to API pricing and tier-based rate limits.
Bottom Line
GPT-6 Astra usage limits are more complicated than a simple fixed number of prompts per day.
For Work and Codex, OpenAI currently estimates approximately:
- Plus: 5–45 local messages per five hours
- Pro $100 / 5X: 25–225
- Pro $200 / 20X: 100–900
- Business Standard: 5–45
Weekly limits may apply on top of the five-hour allowance.
GPT-6 Pro in standard Chat uses a separate system, with 50 messages per week on Pro $100 and 200 messages per week on Pro $200 under OpenAI’s current limits.
The amount of Astra usage a task consumes depends heavily on task complexity, reasoning effort, input size, output size and tool use. That is why the published ranges should be treated as estimates rather than guaranteed message counts.
For most users, the simplest way to avoid unexpected limits is to reserve Astra for difficult work, use lighter models for routine tasks and check Settings → Usage before starting a large job.
Because OpenAI is actively adjusting Astra availability, subscription options and usage systems during the rollout, limits may change. Check the official OpenAI Astra usage guide for the latest first-party information.
