@Dev4YM

@flavioAd FR

Getting rate limited is fine Killing the task halfway through is not claude code should just let the task finish like codex does

I hope that's just not true... x.com/Dev4YM/status/…

@Djw_learn @LuminaXspace FR x.com/Dev4YM/status/…

I really don’t understand why they would do that. Claude Work and ChatGPT Work are basically the same interface, just simplified for people who aren’t software engineers. You could take any capable model today, give it the right tools and skills, and recreate all of what these work interfaces do without training an entirely new model.

🚨 Cursor’s next major AI product has leaked Cursor is developing a general purpose agent codenamed “Sand”: • Designed for people outside software development • Can reportedly respond to emails and text messages • Able to organise spreadsheets and documents • Also supports engineering and technical workflows • Internally released to Cursor employees in late June • Development reportedly began in April Sand would place Cursor directly against Claude Cowork and ChatGPT Work. Cursor has reportedly not yet decided whether to release it publicly yet. Is this something you would use?

@LuminaXspace C’mon, Cursor uh. We don’t need another company trying to do everything. Cursor just fix the core SWE experience and agentic models first.. There’s still plenty to improve before jumping into new industries..

@DeepakNesss @sivabuilds Advise you to use OpenRouter for any API. One unified api/model/provider marketplace for all AI models with just one Credits Wallet..

@sivabuilds Currently, these are the subscriptions I have. I keep subscribing to different one each time. Currently, I use Claude for frontend design and Cursor for everything else. Btw, I keep everything updated on this page: deepakness.com/ai/

I have a Routine set up in Claude Code web that sends "Hi" to the agent every 5 hours, starting around 3 AM, so I can better utilize Claude limits. After returning from the gym, I work for an hour starting at 7 AM. Limits reset after about 1–1.5 hours. Then I take a few-hour break and start working again around 10–11 AM; limits reset again in 2–3 hours. Then lunch break, and so on. This schedule works for me, but you can set up any custom cron job expression that best fits your workflow and helps you utilize your Claude limits.

@DeepakNesss Wait. Wait.. That's actually smart 😂

@ProtonDrive u already forgot google.com...

When someone says “no company can beat the Google ecosystem” show them this.

@thegenioo lol

Yeah I am definitely cooked 🙂

@thegenioo and how many is the felt decrease?

@Dev4YM their claim is 10% increase

I really don't know if removing the 5-hour limit is good or bad for us as users. I've got plenty of ideas in my head that argue it's a good thing, but something tells me it's bad.

@thegenioo This is just an assumption and I don't think OpenAI would do it, on top all of that it's just temporary and will end btw (Not if OpenAI changed its mind). For me I don't see any change (increase/decrease).

@Dev4YM it might have decreased overall limit

@sama your new sol became the worst in searching after it was the best.... :(

@ParthJadhav8 @align_all @BoxBox_Club @floyidism @Kamal_K2 @arkade_club But After is more cleaner tho..

@align_all @BoxBox_Club @floyidism @Kamal_K2 @arkade_club I like before so much 🥺

Before / After @BoxBox_Club

@thegenioo @cursor_ai Oh good lol! (Sub to Pro+ btw will be much better)

I am Gromposer Maxxing right now in Cursor

@DeepakNesss Then what's the benefit. I would delegate all the heavy/real work to grok and limit composer for exploration/tracing. For me Idc its costly, good thing that it's better at long-horizon tasks and delegation than composer, and the pricing difference is kinda unnoticeable.

@Dev4YM grok 4.5 is costly. one thing you can do is, include this line below all your propmpts: Delegate all heavy works to Composer 2.5

Used 1,002,549,305 tokens on my Cursor Pro+ ($60/mo) plan in the last 17 days, and I still have ~30% of the monthly usage left. 😍 This the best value-for-money AI subscription right now. 1B tokens 🤯

@donerkebab7 @thegenioo Correct. But when I said, “Grok 4.5 is more expensive,” I didn’t mean that it consumes more tokens. Instead, I meant that its pricing is higher.

@Dev4YM @thegenioo False. Grok 4.5 uses much less tokens to solve problems, so you will pay less to carry out the same work.

If you were already using Composer 2.5 (Fast), you must be switching to Grok 4.5 (Fast) in Cursor The rate difference is very small, where Grok 4.5 is even more tokens efficient, more intelligent, and there isn't an aggressive bump in the cost You will feel more noticeable difference while switching from standard Composer 2.5 to Grok 4.5 considering you were on $20 plan

@thegenioo Yes, you’re right, maybe it’s even better. However, Grok 4.5 still more expensive. The difference might seem small at first, but with high token usage, it becomes much more noticeable. Personally, even with relatively low usage, I still consider the price difference significant.

@Dev4YM I am not even showing promo pricing, I am showing actual model pricing which competes with Composer 2.5 without any discount So what work you were getting done by Composer 2.5, you are going to get almost somewhat similar on Grok 4.5 even after promo

@Zortosdev bruh it's temporary.. x.com/thsottiaux/sta…

OOH HELL YEAH no more 5 hour limits