Squşyonka Enjoyer

[Teortaxes▶️ (DeepSeek 推特🐋铁粉 2023 – ∞) @teortaxesTex]

Top 3 labs: Anthropic, OpenAI, DeepSeek. Maybe other Chinese labs qualify in taste but not in ambition.
Google, xAI aren't worth the GPUs they've bought, so they're renting it out to Anthropic. Close to Amazon tier, actually. Embarrassing skill issue.

[Teortaxes▶️ (DeepSeek 推特🐋铁粉 2023 – ∞) @teortaxesTex]

I've become numb to the injustice. Not just gigabrained EAs on the frontier, but some lowly scum in Trump's entourage are getting filthy rich off this repeated pump and dump. Do you think Trump talks to Iranians? No, nobody believes this. It's all bad faith insider trading.

[Squşyonka Enjoyer @DeviousScorpios]

@E_redfieldc C-UAS
USV

[DeathInFire @Death1nFire]

@teortaxesTex This is how I feel when I see datacenters, chip fabs and other startegic facilities built with 0 defenses whatsoever.
They are going to catch hypersonic shells.
But spending 15 billion on a battleship thta would die in 1 hour is apparently a better use of money

[Squşyonka Enjoyer @DeviousScorpios]

@teortaxesTex 1.37... will decrease even further over the next years

[Zoso 🇺🇳(🇮🇶) @MesoAnglic]

I am very proud to announce that Wasaburo Oishi's 1926 report has been fully published for the first time on the open Internet, in its original Esperanto and in a translation I wrote into English!

The earliest documentation of the jet stream is now avaliable publicly!

[Squşyonka Enjoyer @DeviousScorpios]

@teortaxesTex

[Teortaxes▶️ (DeepSeek 推特🐋铁粉 2023 – ∞) @teortaxesTex]

> be German
> get brain-fried by edgy Continental philosophy
> back a charismatic Führer promising national rebirth and superiority through conflict with the world
> support his movement that memes the country into a war of aggression motivated by a kooky doctrine about the special historical importance of Jews
> actively contribute to the war effort
> proceed to lose the war
> flee to Buenos Aires, Argentina
> be Adolf Eichmann
> picture unrelated

[犬 @not__dog__]

“This is a forbidden zone.”

[Squşyonka Enjoyer @DeviousScorpios]

@MesoAnglic Some of the melodies are actually based on the station names. For instance, the chime at Qara Qarayev station is from Qara Qarayev's Yeddi Gözəl ballet, and Koroğlu station plays the Koroğlu overture

[Bora Tas @Tas1Bora]

@Alfa_Particle This topic is a good indicator to gauge who missed the last 15 years

[OK Then @okaythenfuture]

@TheMindScourge Thick vs Thin time.

[DeepSeek @deepseek_ai]

🚀 DeepSeek-V4 Preview is officially live & open-sourced! Welcome to the era of cost-effective 1M context length.

🔹 DeepSeek-V4-Pro: 1.6T total / 49B active params. Performance rivaling the world's top closed-source models.
🔹 DeepSeek-V4-Flash: 284B total / 13B active params. Your fast, efficient, and economical choice.

Try it now at https://t.co/GCdiMzk1Dl via Expert Mode / Instant Mode. API is updated & available today!

📄 Tech Report: https://t.co/drlDrxkYtp
🤗 Open Weights: https://t.co/T13Y8i7SDM

1/n

[Taelin @VictorTaelin]

update: anthropic essentially told me my company isn't big enough to be part of the glasswing partnership so I'll make it be

[slinky vagabond 🌲🔩☕️🦉🌻 @83dollaroring]

you could probably write an intersectional mechanical engineering + political sociology dissertation on american versus russian/soviet small arms design

[Squşyonka Enjoyer @DeviousScorpios]

@teortaxesTex @MazMHussain Another Teor banger

(It’s been a while. I hope you’ve been doing well)

[Teortaxes▶️ (DeepSeek 推特🐋铁粉 2023 – ∞) @teortaxesTex]

The hero decels need, not the hero they deserve.

[Cᴀʟɪʙʀᴇ Oʙsᴄᴜʀᴀ @CalibreObscura]

A VKS-level lack of planning for enemy capabilities

[outside five sigma @jwt0625]

been thinking about both economically and culturally how do you enable small players to innovate and move fast on insanely large scale systems and projects such as infrastructure, as old and big players are slow and there are so many things far from optimized, and it really comes down to having good metrics and observability to build good simulations, and make games out of the simulations to let the kids play. Both actual kids and big kids, both digital games and real physical games.

sadly the most hard part to simulate is the human interactions. Figuring out and optimizing the tech is simple compared to actually convincing people to do it together.

[outside five sigma @jwt0625]

@teortaxesTex good engineers are harder to convince than good investors