𝗭𝗲𝗻 𝗠𝗮𝗴𝗻𝗲𝘁𝘀

I take back my disparagement of @0xSero 's AI rig for the janky PVC trim holding his GPUs up. His GPUs are clearly living the good life. Sitting on nice hardwood floors, breathing fresh human air, eating wet food every day. My shit's in the basement gulag, chained by trip hazard wires, coils whimpering unheard. Ty @LLMJunky for helping my nubass build it.

@MatthewBerman @LLMJunky @TheAhmadOsman @0xSero @ivanfioravanti @NVIDIAAI Yup

@LLMJunky @TheAhmadOsman @0xSero @ivanfioravanti @ZenMagnets @NVIDIAAI 96gb each?

Finally proud to announce that I've joined the GPU Minor Leagues. 2 x RTX 6000 Pro. I have six months to pay off the second GPU lol. You are all TERRIBLE influences.

@RedEagleAnon @nickshirleyy Would check out Mango House. There are some legit businesses, but also some weird fake medical offices that don't accept any clients. maps.app.goo.gl/ruZP3LtfaMhgKt…

Day 3 asking @nickshirleyy to help to discover fraud in COLORADO!!!!!! This deep blue sanctuary city state is hurting the American citizens so much… We call it Colofornia…. Or Calirado…… whatever…. Drop the best CO nickname below!!!! 👇

where the heck is deepseek v4

@sweinoid @LLMJunky @TheAhmadOsman @0xSero @ivanfioravanti @NVIDIAAI It's time. Get in here Will discord.gg/pA2DsV82g

@LLMJunky @TheAhmadOsman @0xSero @ivanfioravanti @ZenMagnets @NVIDIAAI Id send you private link to 6000 owners discord but cant seem to dm. Youll love the cards

@LLMJunky @Snixtp @TheAhmadOsman @0xSero @ivanfioravanti @NVIDIAAI You bought 128gb ECC DDR5 for $1k?! Damn. I thought I was doing alright for $1600

@Snixtp @TheAhmadOsman @0xSero @ivanfioravanti @ZenMagnets @NVIDIAAI i paid 1k for the ram, and that hurt

Minimax M2.7 vs Minimax M2.5 in the voxel pagoda test. Good work @MiniMax_AI Comparison with other SOTA models here: x.com/ZenMagnets/sta…

@grok Two additions: Minimax M2.7 and Composer 2

Qwen3.5-397B vs GPT-5.3-Codex XHigh vs Grok 4.2. Qwen3.5 is solid though not sota tier. Codex knocked it out of the park, but was the only one that couldn't 1-shot it - first two tries ended up in bugs, which is not typical for what has become my goto coder. @Grok 4.2 has a high voxel count but is disappointingly dull for Feb 2026, even compared to smaller open models like M2.5. For all tests: Unless they auto-rotate, the perspective changes are from mouse drag.

Voxel Pagoda Comparison: Gemini 3.1 Pro vs Gemini 3 Flash vs Opus 4.6 vs Kimi k2.5. See 🧵 for GLM-5 vs GPT-5.3-Codex XHigh vs GLM 5 vs Sonnet 4.6 vs Minimax M2.5 vs Qwen3.5-397B vs Grok 4.2 Gemini 3.1 Pro is very strong, and has a high voxel count. Opus 4.6 seems to have the best taste. At least in these 1 shot comps. K2.5 outputs on par with what previous version of Gemini Pro could. Flash is lazy. Prompt: "Design and create a very creative, elaborate, and detailed voxel art scene of a pagoda in a beautiful garden with trees, including some cherry blossoms. Make the scene impressive and varied and use colorful voxels. Use whatever libraries to get this done but make sure I can paste it all into a single HTML file and open it in Chrome."

@LLMJunky @Snixtp @TheAhmadOsman @0xSero @ivanfioravanti @NVIDIAAI Bet you'll be able to run the NVFP4 of this reaaal soon!

@Snixtp @TheAhmadOsman @0xSero @ivanfioravanti @ZenMagnets @NVIDIAAI just 400gb or so short from running something actually good lol

@LLMJunky @TheAhmadOsman @0xSero @ivanfioravanti @NVIDIAAI Welcome 🙏

@keennay Damn, good thinking 😅

@ZenMagnets missed placement opportunity for Claude’s logo

If all we need to fix what's going on in the Strait of Hormuz is a finger, why haven't we just deployed the Canadian Curling team?

@LLMJunky @moltbook Unlikely to be true

Okay I knew the guy who sold @moltbook was a multi millionaire after this deal, but hold on. WHAT? $200 MILLION DOLLARS? There's absolutely no way. Wow.

@TheEvryDayChris @Tesla @nvidia My favorite talk so far (and also my giveaway entry submission!)

Nice to hear @tesla at @nvidia

Excited to see @aelluswamy at @NVIDIAGTC

@_Curious_Human_ Nice

@ZenMagnets It's 72 GB VRAM, it is a detail, but important. Running like that:

Arguing with my Qwen3.5 397B running locally on 3x3090 and 128 GB RAM at 20 tps. It says it can't run locally on my HW :-D :-D.

@hussein_builder @NVIDIAAIDev @karpathy @DellTech @grok $96k

@NVIDIAAIDev @karpathy @DellTech @grok what is that thing priced at sticker

🙌 Andrej Karpathy’s lab has received the first DGX Station GB300 -- a Dell Pro Max with GB300. 💚 We can't wait to see what you’ll create @karpathy! 🔗 #dgx-station" target="_blank" rel="nofollow noopener">blogs.nvidia.com/blog/gtc-2026-… @DellTech

@KyleClark I'm not an Evangelical Christian. Atheist AF. Yet I can see how permanently mutilating children can be accurately be described as demonic. Be more reasonable Kyle

Evangelical Christians have successfully placed two anti-trans measures on the November ballot to "stop the transgender plague in Colorado." The initiatives sidestep the Democrat-controlled legislature, which a ballot measure organizer says is inhabited by "actual demons."

@GuntherEagleman But I want the money returned first. So go into their accounts and claw everything back, and liquidate all of their assets, and make them work off the remainder in prison before being returned home.

🚨BREAKING: The US House just PASSED legislation to DEPORT migrants who commit fraud against America, welfare scams, fake claims, document fraud, you name it. 186 Dems voted NO. 186 traitors who’d rather protect foreign fraudsters than American taxpayers.

@LLMJunky @Teslanaut @0xSero Here for you bro

@ZenMagnets @Teslanaut @0xSero i'm gonna be the most annoying person on the planet after i get this setup lol. just a heads up. i need you.

Funniest thing I read all day.

@LLMJunky @Teslanaut @0xSero I think 2 is the best value right now. I expect you'll be able to run Minimax M2.7 NVFP4 when it's out. It's been hell trying to get Qwen3.5 397B to perform properly.

@Teslanaut @ZenMagnets @0xSero i got 2 before the price increase