pzo

> The real loser in all this is obviously Europe, which has nothing competitive once the Chinese models are gone.

Those models are open source / open weights. Europeans can continue with Chinese models as a base same like Cursor used Kimi K2.5 a base for their composer 2.0 model or SWE-2 from cognition is post-trained Kimi K3

> I'm Polish, out of my 8 grandfathers

You probably mean great grand parents. You can only have 2 grandfathers.

I recently was hunting for some storage on my ssd and noticed had 14GB of tiny files in codex in /.staging/marketplace. They really don't care about our drives. Once they fix one things they break another stuff.

And big usage of energy is actually heating home so heating water

A lot of energy is used for heating, heating water is more efficient with fossil fuel than with electricity - because for electricity you have transmission loss.

Super Intelligence name change is most likely about enrichment - there were reports about sales of .si (slovenia) increased significantly many months ahead of annoucement - most likely insiders.

tested in polish and unless you speak very loud, clear and slow is not that good, parakeet definitely better.

there is not information that's exactly the same bacteria and that drugs work. It might be novel one or modified in the lab. Even for flu we don't have general vaccine that works.

He quit google so that he could speak openly about this topic. He explained in interview how he even got into google before - sold his company to google because he has adult kid that is handicapped and he is the only provider and be in this life forever - a fair choice IMHO.

AFAIK he didn't expect this will develop that fast. The biggest issue is not that technology is dangerous but that we develop it a break-the-neck speed.

Geoffrey Hinton (the real AI godfather and Nobel price winner), who was LeCun supervisor, called his student (LeCun) the crazy one in one interview AFAIK

you can pretty much say the same about many countries like italy, greece or even developing one. But we should be comparing to last 1-2 decades rather than 1-2 centuries. And in my opinion precovid era in poland was better than current one.

golden age was precovid we are passed the peak even thought we got into G20 I suspect GDP increased a lot of because inflation.

Been living a lot in SE Asia in the last 12 years but ever yeah in poland for few months. There has been really big inflation in the last 2-3 years and everybody notice it.

Been twice this year in Italy, one time in alps and last week in puglia and there is a reason so many Poles go there for holiday - its cheaper or similar price but better than taking a holiday in poland.

Wish they explained more. Even though I'm mostly native iOS its seems react native is better stack for AI and iteration (hot refresh etc) - the main reason pretty much all AI vibecoding app were implemented via expo/react native.

I also believed that finally this year react native got mature enough with improved tooling and performance to the point that it started to being 'boring' technology.

There is a reason that South East Asia is popular for digital nomads and for retired expats. Thailand and Vietnam have still very affordable places if you avoid tourist traps. I would recommend first trying some easier destination like chiang mai or danang (if first time in asia), staying for 2 months (vacation/workation) and seeing if you still like it.

yeah some people are very orthodox and think having a partner, kids, pets, mortgage is a must. The easiest exceptions is being a monk or priest and they are fine and society accept it.

How then about monks or priests? I wouldn't clasify them as all of them being sad or having a fundamental need to be loved. Also loved can have different meaning - mostly people think about family and partners. There is also community.

probably less - I sometimes manage to burn like most of my 5h credits on Claude 5x Max plan in 2h.

The annoying things is that you gave bigger tasks and is such session finished maybe 70% and after 5h pool resets cache will expire so that if you "please continue" it will wipe out again a lot of your 5h pool)

This is a significant list but unfortunately not load-bearing enough.

This simplify integrations. Right now you have to maintain your own billing credits for user and even maintain user database, abuse detection, planning when you have to top up your openrouter accounts etc. Lots of hassle and this is a reason why many developers charge e.g. more in subscriptions the end user.

It's more user and developer friendly when you tell your user:

- you don't have to create new account just log in to your openrouter account

- bring your own key

- you don't have to pay any subscription. we charge small markup fee (e.g. 5-10%) over your usage inside our app.

> Set your model and effort level before you start. Changing either one mid-conversation can bust your prompt cache, which can increase token cost.

I know we supposed to do this but is there any particular reason why such things cannot be supported? I thought its running on same model just different settings like reasoning. This would be super useful.

Wouldn’t that invalidate kv cache and you wouldn’t benefit from much cheaper cache read?

I doubt they will match old cache read pricing- that’s most important in agentic coding.

no, they said the model is still 1.5T size. Next one 4.7 is supposed to be bigger.

highest selling is not same as cheap - apple also have the highest selling iphones among smartphones and they are still pretty much the most expensive (and usually not the best either these days).

>> I think they will pull ahead with cheaper tokens similar intelligence

they just increased cache read from 0.30 to 0.50 - this has the biggest impact on agentic coding. Elon companies have the most expensive everything:

xAI sub: $30 when other starts at $20, pro like sub for $300 where other charge $200.

Expensive electric cars, powerwalls, solar roofs when competetive products/better are cheaper.

Seems the cache read pricing almost doubled from $0.30 in Grok 4.5 to $0.50 in Grok 4.6.

In my experience in heavy coding sessions most pricing is just cache read and cache write like 80% of my token bill.

there is another total solar eclipse next year in 2027 in south of spain if the first time not lucky

if you want to attract more userbase that's probably the least meaningful copy-text someone can provide. Not only doesn't even in one sentence what sonic pi v5 is but also doesn't explain whats difference between v4 and v5 - I mean what it means more friendly/accessible/fun really?

I worked around similar problem just working from coworking space and/or traveling.

For cowork you don’t often have to commute as much (more options around), can arrive late, don’t have to dress to nice. For some you can still change your homebase at least for few months temporarily.

But you can still have physical coworkers (even if you don’t work with them), have chitchat not relate to work, go to lunch together. You can also have routine and separate your professional life from personal one.

Short sellers earn the most when company they bet against is crashing.

On the other hand investors benefit when they competitors are crashing.

For many labor is just expensive cost running company - I can see how some people salivate