Maحmoud Ismaعil

الحَمْدُ للهْ حتى يبـلـغ الحَمْدُ مـنـتـهاهـ ... اللهم ولك والحمد

[Maحmoud Ismaعil @XGenerationz]

This already big problem in opus; lazy.
About codex it’s normal from long time, never give up… also cursor in auto mode, maybe using GPT

[Maحmoud Ismaعil @XGenerationz]

Today, I fixed the issue with my card on the bank's side. My bank had blocked OpenAI by mistake, so it's time to get back to my friend Codex. I'll keep Cursor (GLM & Kimi are already long-term), but let's be honest: Fable got nerfed, and Opus got better. I think it's back.

[Maحmoud Ismaعil @XGenerationz]

Is it really worth the price?
I highly doubt there's an urgent need for such pricey tools, or for dev. to rely solely on Claude. There are better alternatives on the market. While Codex is a solid option (despite its lack of plans), like Cursor are currently the ultimate choice.

[Maحmoud Ismaعil @XGenerationz]

@ravikiran_dev7 Sure cursor!

[Maحmoud Ismaعil @XGenerationz]

As per i have a problem with billing issues with ChatGPT, and big issue with claude as models only, with non reasonable, i tested deeply Cursor, its really impressive.
Specially in Auto Mode, it’s really excited. Easy and reliable.
Cursor with GLM subscriptions; perfect for me.

[Maحmoud Ismaعil @XGenerationz]

GitHub Copilot officially offers the Kimi K2.7 Code model (by Moonshot AI) as a selectable option in its model picker. It is the first open-weight model available in Copilot, providing a lower-cost, highly competitive alternative for coding workflows

[Maحmoud Ismaعil @XGenerationz]

@ericzakariasson No needed, cursor already stronger 💪

[Maحmoud Ismaعil @XGenerationz]

Be careful, because I'm dangerous now 😂

[Maحmoud Ismaعil @XGenerationz]

@Zai_org Already use it 💪, keep up 🔝

[Robin Ebers · AI for Small Business @robinebers]

the real question is

GLM 5.2 vs Sonnet 5

for coding/design → GLM 5.2 Max

for everything else → Sonnet 5

general knowledge will still be better with the Claude models

but for coding GLM 5.2 beats it

[Maحmoud Ismaعil @XGenerationz]

@bridgemindai with removed sonnet limit :))

[Akhilesh Mishra @livingdevops]

Vibe coders everyday struggle in 15 seconds.

[Maحmoud Ismaعil @XGenerationz]

"50% Discount for the U.S Government: An agreement between Anthropic and U.S Government to use Claude at half price."

[Maحmoud Ismaعil @XGenerationz]

Cursor is smarter than Claude Code; in the same case, Claude adds skip/hide failures ... Cursor had a different opinion.

[Ben Cera @Bencera]

China will crash the US stock market this year.

They're dumping SOTA open-source models on the US, nearly as good as the frontier labs, basically free.

Nobody can refuse cutting a $100M AI bill down to $5M.

Once that deal is on the table for everyone, Anthropic and OpenAI can't justify their prices, or their valuations.

The bubble isn't AI. It's the price of AI. And it's about to pop.

[Maحmoud Ismaعil @XGenerationz]

@kimmonismus no mobile app

[Maحmoud Ismaعil @XGenerationz]

@sama It speaks volumes about OpenAI's commitment to its users when support inquiries are met with such clarity, patience, and genuine care. In a world where automated or non-replies have become the norm, It reminds me why I chose to be part of this platform in the first place.

[Rohan Paul @rohanpaul_ai]

👀 This is from an ex-Meta PM.

[Maحmoud Ismaعil @XGenerationz]

@jietang For example ... Deep Source requires strict adherence to the codebase and cannot deviate from it. There will still be a problem... The only one who followed that, even if the code was refactored properly, is GPT. It adheres completely to what is required and fixes the issues.

[Maحmoud Ismaعil @XGenerationz]

@jietang Sometimes... The model deviates from the desired path... Even if it is made clear that this is prohibited. All models... Except for GPT 5.5, do that ... And the matter is in solving problems, not in escaping.