Vaz

295 posts

Vaz banner
Vaz

Vaz

@vaz_devs

Web dev building a social media UGC content creation and automation platform. Documenting the journey in public

Katılım Ağustos 2025
74 Takip Edilen19 Takipçiler
Vaz retweetledi
Tibo
Tibo@thsottiaux·
Don't just reset Codex rate limits for fun, it costs money. Don't just reset Codex rate limits for fun, it costs money. ... but the vibes are good ... I have reset Codex rate limits for ALL paid plans to celebrate a good week and allow everyone to build more with GPT-5.5. Enjoy
English
1.5K
764
17.2K
1.3M
Vaz retweetledi
Vivo
Vivo@vivoplt·
Claude: I’m ready to build your project. Just say go. Me: Go. Claude: Daily limit reached.
English
79
91
1.5K
26.2K
Vaz
Vaz@vaz_devs·
Codex with browser use is really a game changer 😍
English
0
0
0
21
Vaz
Vaz@vaz_devs·
@0xSero So how to get the browser tab in Codex?
English
0
0
0
793
Vaz
Vaz@vaz_devs·
@vercel Thankfully I don’t use this trash service 🙌
English
0
0
0
15
Vercel
Vercel@vercel·
We’ve identified a security incident that involved unauthorized access to certain internal Vercel systems, impacting a limited subset of customers. Please see our security bulletin: vercel.com/kb/bulletin/ve…
English
696
2.4K
8.8K
10.6M
Vaz
Vaz@vaz_devs·
Opus 4.7 is unusable. Can't build anything off of it with the $20 subscription. I'm done with @AnthropicAI ❌ Any alternative recommendation? Let me know!!
English
0
0
0
12
keshav
keshav@kshvbgde·
opus 4.7 hit the limits with just 3 prompts on the $20 plan
English
88
79
3.9K
181.7K
BridgeMind
BridgeMind@bridgemindai·
Claude Opus 4.7 is so good I just bought two Max subscriptions. After vibe coding for over 12 hours yesterday I can confirm this is the best model in the world. It burns tokens fast. But the coding output is unmatched. Two $200/month Max plans so I never hit a rate limit again.
BridgeMind tweet media
English
207
21
699
127.1K
Vaz
Vaz@vaz_devs·
1 prompt of Opus 4.7 literally consumed 30% of my limits. All I've asked was an opinion on how to improve an UI component, it didn't produce me any code!! I'm simply going to cancel my sub, or I'm stuck using Sonnet 4.6
English
1
0
0
644
Vaz
Vaz@vaz_devs·
@theo Codex 5.3 sometimes really acts dumb and forgets the previous instruction
English
0
0
0
67
Theo - t3.gg
Theo - t3.gg@theo·
Serious question. Has anyone ever noticed meaningful regressions in Codex/OpenAI models? I feel like we talk about this a lot w/ Anthropic but I've never seen a similar discussion with OAI.
English
306
11
1.7K
142.1K
Vaz retweetledi
Thomas Gauvin
Thomas Gauvin@thomasgauvin·
The wait is over. Cloudflare Email Service is now in public beta 📧 Send and receive emails directly from Workers or REST API with global delivery on Cloudflare's network And just in time for you to build email agents with the Agents SDK!
Thomas Gauvin tweet media
English
181
313
3.7K
273.6K
Vaz
Vaz@vaz_devs·
@romxdev Vue fanbase, unite!
GIF
Español
0
0
1
9
Roman
Roman@romxdev·
Unpopular opinion: React completely destroyed web development We spent a decade learning a highly complex library just to reinvent server-side rendering all over again
English
120
57
1.3K
174.1K
Vaz
Vaz@vaz_devs·
@cesaralvarezll Greediness led to this. They deserve it 🤷‍♂️
English
1
0
1
282
Vaz
Vaz@vaz_devs·
@rsuyoy Probably you forgot they had x2 promo and it ended. It's still much more better than Claude tho.
English
0
0
2
366
Yousr
Yousr@rsuyoy·
Why is everyone okay with OpenAI cutting Codex limits by a huge amount and not publicly communicating it? It’s shocking, the limits are now on par with Claude’s code if not stricter, and there’s no formal communication on it.
English
173
27
821
81K
thethiny 🐰🍉
thethiny 🐰🍉@thethiny·
I tried running the same task (an average task for me) against both 5.4 high and opus 4.6 with high effort. Codex used 30% of the 5 hour limit and 7% of the weekly limit. Claude used 10% of the 5 hour limit and 2% weekly. The results? Both did the same job but codex did more stuff than I asked (but were correct). So I think I'll favor usage more than I'll favor slightly better quality. Thanks for your insight and the link 🙏🏻
English
1
0
1
74
Vaz
Vaz@vaz_devs·
So apparently rewinding to a previous message in the new Claude Code just hits your limits. Seems like it forgot its whole context, I still had like ~20% limits remaining. Rewinded a message (which didn't work at all) and then asked a small fix and *poof* limits hit.
Vaz tweet media
English
1
0
1
292
Vaz
Vaz@vaz_devs·
@thethiny Speaking for myself: GPT 5.4 XHigh is great if not better than Opus 4.6 Switching between 5.4 and 5.3-Codex depending on the tasks, to save up tokens. You might want to take a look at this, to have the best experience with codex: github.com/code-yeongyu/o…
English
1
0
1
45
thethiny 🐰🍉
thethiny 🐰🍉@thethiny·
@vaz_devs Is codex on the same level? It's too restrictive and does too many things wrong. It even indents lines incorrectly, and pastes code blocks at the beginning of files.
English
1
0
1
25