Tarun

building AI agents · backend systems · occasionally yelling about football

[Tarun @tarunfcb]

Codex in the last 3 days has been a nightmare. Hallucination after hallucination, writing so much unwanted code, creating things that are not at all intended, taking too much time and burning tokens unnecessarily. Especially
/goal.

@sama something is wrong. It burned 50% of my Pro plan tokens for the week in 1 night creating a PR of size 104k lines and when asked about the PR : it said it’s useless and cleared the whole PR and closed it.

[Elon Musk @elonmusk]

Unfortunately, we are being blocked by certain oligarchs in order to maintain their monopolistic chokehold on the Indian people. You can guess who they are …

This is a crime against the people of India!

[Tarun @tarunfcb]

@align_all Engagement farmer.

[Tarun @tarunfcb]

@hany99dev Opus5.5

[Tarun @tarunfcb]

Opus5.5 feels really efficient and faster as well 👏

[Tarun @tarunfcb]

@johnternus @Apple Siri is still so painfully bad 😭

I genuinely can’t tell what’s been improved, and at this point I’m not even sure where the “AI” part is supposed to be.

[Tarun @tarunfcb]

@thsottiaux /reset 😁

[Claude @claudeai]

Introducing Claude Opus 5.5, the first model in our new Claude 5.5 family.

It performs at the level of Claude Fable 5.1 for most tasks, and costs 40% less to run than Opus 5.

[Tarun @tarunfcb]

Siri is so dumb 🥲

[Tarun @tarunfcb]

When I forget everything and really support Real Madrid against those clowns Atletico - RM idiots stink all over the pitch.
What a shit team man 🥲

[Tarun @tarunfcb]

Where’s the road runner Valverde?

[Fabrizio Romano @FabrizioRomano]

🚨🇦🇷 Barcelona president Laporta on Julián Álvarez: “Atlético have showed true colors. They acted as though they were the offended party — they had no way of justifying their attitude. We did what is normally done — as there were talks between the clubs, then I personally spoke with Gil Marín”.

“He told me Julián wasn’t coming because they didn’t have an alternative… and then Atlético reacted inappropriately. I do not know what caused it. I can imagine why, but I have no proof”.

“We were hearing reports that they would sell to anyone in the world except Barça. And that’s the case… they were actually ready to sell Julián to any club in the world, but not to us”.

“It’s already the past, we’re now very happy with our squad”.

[Diogo Almeida @CompleteSkeptic]

After co-inventing ChatGPT, I kept asking myself: why have superhuman chat models not led to AGI?

I’ve spent the last 2 years in stealth building a new way to train models (RLCD), and a new type of frontier AI model that we are releasing today: Jev

• 20-200x faster
• 40-400x cheaper (w/ output tokens free)
• Frontier composable intelligence optimized for decisions

AFAICT the shortest path to AI-based economic revolution

[OpenAI @OpenAI]

We’re sharing a solution to the Navier-Stokes Millennium Prize Problem, one of the deepest problems at the frontier of mathematics.

The proof was produced by a group of agents, using an OpenAI next-generation model significantly more capable than GPT-6 Astra.

The problem concerns whether the description of smooth three-dimensional fluid motion modeled by the Navier-Stokes equations can break down. It has remained unresolved for roughly 90 years.

[Tarun @tarunfcb]

GPT-6 Astra is easily the best model I’ve seen at browser use. Incredibly accurate, seemingly unstoppable. It just keeps going and keeps getting it right.

[Tarun @tarunfcb]

It’s here!!! 🔥

Got that crazy excitement to try out new models for Astra again.

Been waiting for this feeling. Let’s fucking gooo! 🚀

[Tarun @tarunfcb]

HOW MANY DAYS WILL THIS ROLLOUT TAKE??
WHERE IS ASTRA??

[Tarun @tarunfcb]

I think its coming!!!
#astra

[SemiAnalysis @SemiAnalysis_]

Recently, we purchased one of each Anthropic/OpenAI subscription plan and randomly ran long horizon coding tasks until we exhausted the weekly limit. It's widely believed that a $200/month plan maxes out at ~$2000/month worth of tokens (assuming API pricing). However, we found that the subscriptions are actually far more generous. (2/4)

[Darren @dtzy_88]

这次 ChatGPT quota nerf，基本已经被社区实锤了，下降了将近 76% 🤬

SemiAnalysis 6 月曾实际测试过各档 ChatGPT 订阅，当时估算出来的 API 等值大约是：

$20 Plus → ~$700/月
$100 Pro 5x → ~$3,500/月
$200 Pro 20x → ~$14,000/月

但最近 Reddit 上几位用户测出来的数据是：
Plus：额度下降约 20%～50%
Pro：部分账户下降 70%+ ！！！

其中一个 Plus 用户的 NerfTrack 数据尤其有意思。

上周 reset 之后，他观察到的并不是“一刀砍到底”：

OpenAI 第一次先砍约 40%，发现没人注意后，又继续砍了 10%，最终相对原额度只剩一半。

另一个 Pro 5x 用户的数据更夸张：
$674/周 → $157/周

直接缩水约 76.7%。

所以 Pro 用户多付 5倍的价格结果却拿到跟 Plus 差不多一样的额度 ？？？

OpenAI 你是真的把用户当猴耍啊 🤬

Tibo 搞不好又要用��� bug 搪塞过去了 🙄

如果你最近感觉 Codex “怎么突然这么容易花完 quota”，你可能真的不是错觉。

Plus / Pro 用户来报个数：

你现在一周实际能跑多少？
我尤其想看看 Pro 用户的数据 👀

SemiAnalysis：
https://t.co/0CMtINxDeP

Reddit：
https://t.co/OYyZE6pk4f

https://t.co/SHcxkG571O

[BridgeMind @bridgemindai]

$200 A MONTH FOR ONE DAY OF USAGE.

OpenAI Pro is cooked. I burned my entire weekly GPT 5.6 Sol limit in a single day of Codex. 0% left. Reset is 5 days away.

This is not the OpenAI I used to know. They went from the best limits in AI to the worst.

Meanwhile SuperGrok Heavy lets me run Grok 4.6 all day and I barely touch 10% of my weekly.

Grok is winning on usage and it is not close.