sschueller

Try Kilocode with deepseek v4 (via API directly to deepseek, much cheaper than via kilo).

I have had very good results and compared to others it just costs pennies.

I use something similar to this https://github.com/ScotterMonk/AgentAutoFlow setup and switch between deepseek v4 to flash depending on task.

Amazon wasn't competing against open and free models that are starting to be good enough running on existing laptops.

OpenAI and Anthropic's moat is filling with cement faster than they can dig.

I don't want to be cynical, but I assume a third party we can trust has verified this model is actually this good?

I would think it would not be Anthropic, out of all the players, that is selling a lie hidden behind "I am sorry, I can't do that; it's too dangerous."

This has been the case for a long time in Switzerland. If you want a mobile phone SIM, you need to provide your ID. Of course easy if you have a national ID...

I don't see the issue here and I am surprised that this wasn't the case in the United States.

However since it is now trivial to communicate encrypted with so many other devices, I don't see the point now. This would have been a "good" thing back in the 90'.

> Wild meat, also known as “bushmeat,” refers to the meat of wildlife species hunted or collected for human consumption. USAID staff developed the Wild Meat Learning Agenda to generate and share evidence to inform efforts to improve wild meat programming and to understand those connections to food security, health, and conservation. The Learning Agenda defines learning questions and associated activities to address those questions.

https://www.wildmeat.org/toolkit/

But when they fail, they fail upwards. These risks are always on the backs of everyone else.

For example, they take risks by using chemicals (that they know) can kill people, by the time people get cancer they are retired, there are no claw-backs.

More recently you commit actual fraud, get convicted but then get a pardon and somehow end up with money again running the next "scam".

These people never take real risks.

DOGE resulted in the death of over 100k people [1] and will result in many people dying of Ebola. It was a huge failure yet the "risk" Elon took is not his.

[1] https://healthpolicy-watch.news/the-human-cost-one-year-afte...

They yet have to formerly or directly call what is happening in Gaza a genocide.

This paper is a propaganda and consent manufacturing device of the state voluntarily or not. It is due to many factors but also access journalism.

Let's just stop the shenanigans and just already have the CIA finance it directly...

You don't even need a vote. It's technically not feasible, it is that far fetched from reality.

The point-to-point travel with starship is the one that urks me the most. It is completely unrealistic and will never happen. They have Zürich, Switzerland as a destination. There is no place in all of Zürich to facility such a thing, you will blow out every window in the city at each launch. Absolutely ridiculous that anyone would take this serious.

Also don't get me started on data centers in space idea...

When the Chinese land on the moon sometime in 2030 and the US still doesn't have a way to get there, will Elon finally reap the consequences for his lies or just the interim NASA admin that gave Space X the contract?

Where will it be listed? I am considering selling all my index ETFs in those markets until the this blows over.

I would be worried about a device called "AR-924 Pager" but no one thinks that is an issue.

I was the first to setup a Meshcore room server in my city. New I get pings from all over the place and it's busy. It seems to be extremely popular in Switzerland for some reason.

True but how many times have people sent someone "let me google that for you".

Some people are inherently lazy and unload their laziness to someone else to do the thinking for them.

I still think sending someone an AI answer is terrible but then again, if you are going to ask me for help, at least make some effort first.

EDIT:

By laziness I mean that there are known places (they know of) with documentation that cover what they need but they don't go there first and not something I have some deep domain knowledge of that would take them a long time to find or figure out.

I would personally still not reply with an AI answer but I am tempted sometimes...

Sorry, but that is grotesque. I don't want a Tesla with Ferrari badging.

Meanwhile Switzerland is being taken to the cleaners. F35s that had a fix cost in contract with Lockheed are no longer fixed cost because the US says so.

Patriot systen permanently delayed and price going up and up. Stop payment resulted in the US pulling from the pre payment for the F35s...

And who was it that put her in that situation? An American Journalist that didn't respect boundaries even after it was made clear to them that this would cause issues for her in China.

> Vice published a profile on Wu that included personal details regarding her sexual orientation, which she had explicitly asked them to keep off the record out of fear of state censorship and government retaliation in China.

I pay for google "Starter" workspace.

Recently I started to get harassed to upgrade. Big button in gmail, large notifications on top of my mail in the mobile app etc. Also two other buttons to get me to turn on AI features I don't need.

I already pay a lot, I don't want to pay double just not to be harassed.

Having buttons to features that I would have to pay extra for is one thing. But having notifications and large buttons to upgrade when I am already a paying customer is harassment.

They probably can many things but I think things like memory timing is something you can't just easily reverse engineer from a blob. You need to test every state that the device can be in and see how the blob responds which is quite difficult.

Before or after the AI collapse of 2026/27. I would say at least $499 without the addition of inflated memory pricing.

China can and not with its military.

I trust my government (Switzerland) way more to do the thing that is right for the people and the law then some private company that has the primary goal of making money. It doesn't mean that governments don't make mistakes but the primary goal is to serve its people.

That is what government is for in a functioning democracy. A functioning government is of the people for the people.

PIX is also better because it gives control back to the central bank (as it was with cash) and not private industry although they are providing the service. The central bank controls what payments are permitted by what laws exist, not some risk management system that has decided that your legal purchase is too risky or some foreign state has applied sanctions against you.

It should be possible to sue Google for damages in such cases. This isnt a network outage or service failure which I would consider part of ToS.

It's insane. They would be better of purchasing hardware and running deepseek locally.

At least the usage data would stay sovereign even if the model was trained somewhere else.

The pricing of deepseek v4 flash is incredible. I have been hammering it with kilo code and end up using only cents per day.

Well Claude still thinks it shouldn't read AGENTS.md [1] so they probably also don't care about agents.txt on a web server...

[1] https://github.com/anthropics/claude-code/issues/6235

Like the hyperloop?...

You can only cool by radiation in space. You may get more energy from the sun but how are you going to get rid of all the heat fast enough?

Nobody seem to care about reality anymore or facts. You may as well put a data center at the bottom of the ocean which would be way easier but no one is doing that either.

In the end in like 10-15 years when others land on the moon and build amazing new things maybe just maybe there will be a realization that playing scifi doesn't produce results.