Yeah, so OpenAI has responded to the criticism. People said they shipped GPT 5 with low token limits just to save money and make more profit. OpenAI didn’t just talk, they took action. Sam Altman himself posted on X about it.
OpenAI is testing 3,000 per week limit for GPT 5 Thinking messages for Plus users
This is a big increase from the limits before. But OpenAI says they wont stop here.
Sam Altman claims that soon, all the model rate limits will be raised “above pre GPT 5 levels.”
Here is how we are prioritizing compute over the next couple of months in light of the increased demand from GPT-5:
— Sam Altman (@sama) August 12, 2025
1. We will first make sure that current paying ChatGPT users get more total usage than they did before GPT-5.
2. We will then prioritize API demand up to the…
This could make GPT 5 one of the cheapest powerful models around. Good news for users, yeah?
But wait, OpenAI is also fixing another problem.
People complained it was confusing when GPT 5 automatically switched between its main “Thinking” model and smaller ones like GPT 5 mini.
You didn’t know which one was answering you.
So what’s new?
OpenAI is testing a UI indicator.
This little thing on screen will show you which model is actually working on your question right now. That should fix most of the confusion people had with GPT 5.
Also, Sam Altman shared some numbers. He said more and more people are using the reasoning models because GPT 5 can send questions to the right model itself.
Check this out:
- Free users using reasoning models went up big time, from less than 1% to 7% daily.
- Plus users jumped even more, from 7% to 24% daily!
- And this number is expected to keep rising.
Because so many people are using it now, having higher rate limits is super important.
Altman even said “tomorrow or Tuesday we expect to share our thinking on how we are going to make capacity tradeoffs over the coming months.”
So maybe we will see how they decide between things like ChatGPT vs the API, old users vs new ones, research vs product, etc. Lets see.
Also Read: Claude Learned ChatGPT Trick – ChatGPT Watch Out!
FAQs:
Q1: How to fix OpenAI rate limit exceeded?
A1: Just wait for your limit to reset, it happens after some time. Or consider upgrading your account plan if you need more.
Q2: What is the limit per project in OpenAI?
A2: The limits are mostly per user or per account, not really per project. Team and Enterprise plans have higher limits.
Q3: What are the limitations of OpenAI?
A3: Models can make mistakes sometimes, forget things from long chats, have usage limits per hour or day, and cost money for heavy API use.
Q4: What is the batch limit in OpenAI?
A4: The batch size for API requests can be limited, it depends on the specific model you are using. Check OpenAI docs for the model you use.
Disclaimer
We share news and updates from official sources and trusted websites. Sometimes details change after publishing. Some products or services we write about might be paid or need subscription. Please check info from the official website before buying or investing. We do not have any conflict of interest with any company mentioned. We are not responsible for any decision you take based on our article. Always do your own research. All names, logos and brands belong to their owners.
