Silfaster

@HedgingNapoleon @spellabee @NoTurkeyLegs @jim_benemann Hey Napoleon, just to be sure, did you read the reference to Trump stating he obliterated Irans enriched uranium before moving the goalposts.

@spellabee @silfaster1 @NoTurkeyLegs @jim_benemann And subsequently Trump told Putin he doesn't need Russia to take control of Iran's enriched uranium. Try to keep up.

I think it’s pretty clear at this point that Biden and his team thoroughly miscalculated how this war would turn out.

@HedgingNapoleon @NoTurkeyLegs @jim_benemann Is that relevant? The fear was a nuke within days or weeks right? But here you go, straight from trumps mouth: "But I also obliterated Iran's nuclear hopes by totally annihilating their enriched uranium."

@silfaster1 @NoTurkeyLegs @jim_benemann Again, where is the reference to the enriched uranium being obliterated?

@HedgingNapoleon @NoTurkeyLegs @jim_benemann ;) whitehouse.gov/articles/2025/…

@silfaster1 @NoTurkeyLegs @jim_benemann You failed to include the direct reference to the obliteration of the enriched uranium.

@HedgingNapoleon @NoTurkeyLegs @jim_benemann "“I can say here that the assessment is that we significantly damaged the nuclear program, and I can also say that we set it back by years, I repeat, years.”" like this one? Or "our bombing campaign obliterated Iran’s ability to create nuclear weapons." this?

@NoTurkeyLegs @jim_benemann Ah yes the CNN sound bite posse is out in full swing today. Maybe go read the official statement again or have someone who understands exactly what it asserts explain it to you.

@minhsmind They have ass to kiss to get their free credits, it's pathetic. I'd rather see some useful tools like corridors keying than unoriginal slop made by someone with the mindset of a 12 y/o.

I don’t get this obsession in AI video space to claim that something would have cost hundreds of millions of dollars to make. I get that VFX is expensive. But at a minimum people are ignoring the cost that goes into great writing. I’m sorry to all my friends in AI video but I watch a LOT of television and film on a weekly basis. And I can confidently say that even the best in AI video is still not touching the level of writing we see in popular television shows or films. Nothing in AI video is touching anything in the nominated lists of Emmys or Oscars let alone the winners of these esteemed awards shows. We need to humble ourselves. So please, spare us the hundred million dollar boasts. I’m incredibly frustrated with this because this narrative is incredibly disrespectful to the cinematic arts that I love and hurts the conversation of bringing AI video forward into the light into a healthy dialogue. It’s so transparently incendiary and not at all additive to the moment. I love AI video, cinema, arts, etc and think it deserves a chance. I also think traditional arts, Hollywood, etc. are incredible and have produced some of the best and meaningful storytelling humanity has ever seen. Respect needs to go both ways. It’s unfortunate so many are anti-AI and fail to see the value of it. They fail to see how empowering and fun it can be. But also, if you’re saying Hollywood is done because AI is here, it belies a fundamental arrogance in you and a disrespect for craft and the hard work of understanding emotional tapestry.

@haibaneinsect I'm happy for the vertically eyed people out there that Imax exists, not entirely sure how it benefts the rest of us.

IMAX having this aspect ratio exclusive to only be in their special little cinemas is anti-art. Make the fact the screen in the IMAX cinema is huge as the selling point, not that you can only see the full authentic vision of the director in that specific cinema.

@murdadesmaeeli @phenn @Longshanks_81 @DanBilzerian research done, blue boosted bot

Dear westerners do research before speaking about my country. The mullah Nazis murdered 32,000 people on January 8th and 9th 2026, machine-gunned many of them, old people, many of them kids. That's three times the number of civilian casualties at the eight-year Iran and Iraq war in the 1980s. What America and Israel are doing are saving Iranian people. This is a military aid operation. Understand, please. nypost.com/2026/01/25/wor…

They want to nuke Iran, now they’re just trying to figure out how to justify it

@CEOofTeslat2 @EricTrump my pleasure bot

@silfaster1 @EricTrump Kindly send me a follow request.🚀👋❤️

🤣🤣 One of the great responses to a reporter in history! JAPANESE REPORTER: Why didn't you tell Japan before the Iran war? PRESIDENT TRUMP: "Why didn't you tell ME about PEARL HARBOR?!" "You believe in surprise much more-so than US!"

@rushicrypto The huge depression, you've never seen anything like it. Mdga

If the last one was named “The Great” Depression, what’s this one going to be called?

Tens of thousands of physics-based enemies, now running on Steam Deck, in our indie game There Are Millions of Goblins. Using the 3D Neighbor grid setup for inter-particle collisions, I realised I'm able to detect the concentration of enemies at any given location. Turns out I could increase gpu performance by about 300%, by not rendering any goblins that are obscured by a horde layer above them. Still work to be done. We can make it even better - but this is an exciting step, and hopefully encouraging for players that worry their hardware couldn't handle it! #gamedev #indiedev

@GianmarcoSoresi Can't wait until he can hook himself up to neuralink permanently to enjoy this pathetic slop

You think he beats off to this stuff?

@justfredimelda @robinhobb Isn't it your task as a publisher / marketing fairy to help out with that? Cutting corners by downgrading a book that has cost 100s of hours or more with an AI slop cover made in minutes is plain disrespect to both author and audience

@robinhobb Because some writers cannot afford to hire an artist/do it themselves.

Will you refuse to buy a book with an AI generated cover providing that everything else is human written? #writingcommunity

@RogueAix @eeelistar I am already running models, that's not even the question. But I give up, seems you are unable to tell me what sub $3k setup can run >30B models. anyway, good luck running agents on mobile phones

@silfaster1 @eeelistar why you counting bruh... haha 4090 is a good spec bro depends on what you wanna run, and imo 4090 is a already good to run 30B am running an 10B on my secdonary pc it works fine

I have a team of 10 AI agents working for me This would cost $462,000-810,000 per year in human salaries Instead it costs ~$2,400 per year (Claude Max sub + minimal API costs) That's a 99.5% cost reduction

@Mio_Mind Keeping the flock of MAGA sheep happy in suffering. He forgets to mention that American gas guzzlers on average have half the MPG.

The state of America right now. They find comfort in other people suffering

@RogueAix @eeelistar I know what I can run, am lucky with a 4090, that's not the question. The question (3rd and final try) is: "what hardware would run a model that's actually useable in production for under $ 3k?"

@silfaster1 @eeelistar there's a github repo name llm checker when you run it, it shows what models you can run you should check it out

@RogueAix @eeelistar It was just a question? I'm seriously curious, what would run a model that's actually useable in production for under $ 3k?

@silfaster1 @eeelistar bro you are just exaggerating

@RogueAix @eeelistar Please tell me what runs at least a 30B model under 3k?

@silfaster1 @eeelistar you are just exaggerating at this point bro with 3k you easily buy good pc and run local llm with better alternative agents than openclaw which doesn't too much ra, and yeah about the cost it would vary across different countries

@RogueAix @eeelistar Please trell me, what's a good LLM that runs locally on a phone? 🤣

@silfaster1 @eeelistar bro you just run both cheap llm and low spec agents on your old phones too

@RogueAix @eeelistar Privacy should be the argument, local LLMs do not save you money. unless you already have a GPU and your electricity is relatively cheap.

@eeelistar why not run locally? you don't have to spend that much