Why did ChatGPT add usage limits and how do the new rate tiers work?

ChatGPT added usage limits because running large language models is incredibly expensive — each query costs OpenAI real money in computing power — and the free tier was losing them millions as user numbers exploded past 300 million weekly active users in early 2025. The new rate system works like a cell phone data plan: you get a certain number of "priority" messages per time window, and once you hit that cap, you're either cut off or bumped down to slower response times depending on your plan. For example, ChatGPT Plus at $20/month gives you 80 messages every 3 hours on GPT-4o, while the free tier limits you to roughly 16 messages in the same window before switching you to the less capable GPT-4o-mini model. Pro users at $200/month get essentially unlimited access, but that's a steep jump most casual users won't stomach. What's sneaky is that longer conversations and file uploads eat into your limit faster because they consume more processing — something OpenAI doesn't exactly advertise upfront. If you're hitting these caps regularly and don't want to pay $200, tools like AI-Mind offer a different approach entirely by generating content without requiring you to craft prompts or manage rate limits. For a deeper dive, see our guide on ChatGPT prompts not working. The real lesson here: free AI isn't sustainable, and we're all going to need to budget for these tools the way we budget for Spotify or Netflix. **Related**: How do I check my ChatGPT usage stats? | What's the difference between GPT-4o and GPT-4o-mini?