The cursor pricing model is completely deceiving subscription users

[edumaxv1995]

The current pricing model of cursor is too bad

1. It is not transparent, claude4 is always too proactive in completing additional functions, which makes it difficult for me to estimate the number of tokens I am using and to complete the work according to plan
   In the last infinite plan, I was told that I could use the calude3.5 model without restrictions because it is not an advanced model. Cursor told me to only apply a $20 quantitative plan in advanced models. As a result, I am currently unable to use any other models except for Auto
   I think the cursor team doesn’t keep their promises and has no credibility. They always make a mess in areas outside of technology.
   The PRO+plan clearly states in the client that I can pay $60 to upgrade my subscription, but it is not reflected in the pricing plan on the cursor official website. This is very bad, it makes me feel that the cursor team can freely modify, adjust and play with their subscribed users. This really, really disappoints me.

[nop_nop]

It is. Once the credits are gone, the auto mode is completely useless, it just uses stupid GPT4, answers useless questions, and is totally unusable. Totally useless, feels cheated, and super annoying.

[evilrat420]

Non-hostile pricing discussion

[Screenshot 2025-07-12 231722] I don’t like complaining, but rolling windows for rate limits that allow you to consistently use Cursor for the duration of the month would have been nice, like a set limit on compute usage per hour or some other timeframe. Claude 4 Sonnets token counting system is completely busted, 2000000 tokens per request on average (in the usage dashboard) while opus uses 150000 or Gemini 400000 on THE SAME PROMPT (tried claude 4 sonnet on other platform and it also did not …

[Recevix]

Yes, exactly as you might suspect, something incredibly frustrating is going on here — and as you can see from the link below, I’m genuinely on the verge of a mental breakdown.

First, our credits run out. Then we’re forcibly shoved into this ridiculous so-called “AutoMode,” and guess what greets us there? GPT-4.1 — a completely useless model that can’t handle even the most basic tasks properly.

I’m not just annoyed — I’m furious. This is beyond frustrating, it’s insulting.

( My Psychology Is Broken ) Request for Removal of Useless GPT-4.1 from AutoMode

Alright folks, welcome to hour three of my thrilling war against AutoMode. I’ve now passed the 200-request mark just to get a simple, microscopic change done. And what does this “state-of-the-art” model keep telling me? “I’m on it, I’ll let you know.” Translation: “I’ll waste more of your time and maybe, just maybe, vomit out 20 lines of half-assed code if I feel generous.” Is this a joke? No—this is an insult to the concept of productivity. I kept trying out of some pathetic hope that it mi…

[nop_nop]

+1。Very infuriating and completely unusable. Has the cursor team ever tested this auto mode? I really want to curse.

[Aleks_Sergeenko]

Absolutely agree, bro. I bought PRO+ and now I regret it because it all ended in just 10 days of regular use!! Back when I had the old plan, the same usage lasted me 28 days!!

[v-machine (mario v)]

Have you degraded the early adopters experience?

It seems that greed has indeed taken over. I had high hopes for the Cursor team and jumped on-board very early. I refused to work with any of the ■■■■■■■■ AI coding agents as extensions for a solid 2 years. Then Cursor changed the game. Adoption for our entire company was due to my team’s ability to showcase the benefits… no elaboration needed… moving on… I know for absolute certainty that my development data was used to assist the dev team because the manner in which I was developing to trul…

[34r7h]

frustrating is not the right word. theft and other illegal activities are more suitable

[Alex_Void]

Im on the Ultra plan but according to my calculations, I need 3 of them per month. My colleagues dont hit the limits when working with Claude Max 20x at all even though its much cheaper than Ultra - especially Ultra x3.

The only thing that wins me over with Cursor is some obvious interface features that for some reason are stubbornly missing from Anthropic’s offering.

But that doesnt change the fact that I’m already gradually shifting tasks over to Claude. I think in a couple of months, I’ll make a complete switch from Cursor given this pricing policy.