“We’re also happy to report we’re rolling out changes this week to make our Pro plan more generous. By default, the Pro plan will now follow an unlimited-with-rate-limits model, and all limits on tool calls will be lifted. Existing users can choose to stay with the “500 request limit” method if they prefer (Dashboard > Settings > Advanced).”
It sounds like opting out is just worse, as you’ll limit yourself. So I assume the upside is that you will be getting your fast requests back (Instead of having “unlimited” slow requests)? Should I opt out the new pricing structure when I never needed more than 500 requests before?
I currently subscribe to both Claude Max and Cursor Pro. Lately, I’ve noticed a trend among AI coding agent providers: they’re increasingly obscuring specific usage thresholds and replacing them with vague language. This seems to allow them to “adjust” access to high-performance models—so while I believe I’m using Claude 4 Opus, it might actually be Claude 3.7 Sonnet or something cheaper.
Lately, I’ve found myself more frequently disappointed with the quality of what are supposed to be the most advanced AI models.
If my suspicion is correct, it may indicate that these companies are reaching a stage where they can no longer afford to burn investor money as freely as before.
============================================
I chose o3 as the agent model but the agent answered that he is GPT-4 lol
Your answer is “it depends”. Remember that Claude 4 models are no longer discounted, so Sonnet 4 thinking costs 2 credits per request. That limits you to 250 Sonnet 4 requests, if this is your fav AI model. And it’s still not available in slow mode.
With new pricing model you can in theory use more of it. How the practice looks we’ll only know with time.