Taíde

@CarlosMito43 @Estadao hahahahahahahahahahahahahahaha, queria saber sua opinião sobre vacinas, sufrágio universal e o formato do planeta.

@Estadao Lula tem bosta nenhuma...Mais 80 milhões de brasileiros ferrados.Vcs entrevistam 2.000 mil pessoas,e acham que e o pensamento de 80 milhões de endividados,que trabalham somente pra pagar contas? Tenho 47 anos,nunca participei de pesquisas e nem conheço ninguém que participou.

PESQUISA ELEITORAL 📊 Lula tem 48,9% e Flávio Bolsonaro, 41,8% no 2º turno após divulgação de áudio a Vorcaro, aponta Atlas/Bloomberg → x.gd/Vktdw

@linuscostta @laurojardim BILHÕES??? FORAM TRILHÕES DE DÓLARES, TRILHÕES!!!

@laurojardim Rapaz. A cúpula do PT, partido que ocupa a Presidência da República e a Cúpula do STF envolvidos em escândalo de 500 BILHÕES. Aí a imprensa solta uma notinha dessa como se não tivesse muito trabalho de informação a sociedade a fazer

O apelido de Zema no Novo após chamar de 'imperdoável' áudio de Flávio a Vorcaro dlvr.it/TSbsqN

65 years ago… news.google.com/newspapers?nid…

@Real_SilverLine @Ego_Ergo_Ergot @CattManii @sudoingX dumb project manager says what?

@Ego_Ergo_Ergot @CattManii @sudoingX Nope, been building for years, sorry. The skill issue is you not understanding how bad what AI is outputting is. Because you don’t know what you don’t know. Ignorance is bliss

i've run a stack of models across a single 3090, a 5090, and a 128GB DGX Spark. exactly three are worth building on. the honest list. the three worth it: > 1. StepFun Step-3.5 Flash, the REAP pruned 121B MoE (Q6, DGX Spark) a 121 billion parameter mixture of experts running on a single desktop box. the most worth-it model in everything i've tested. > 2. Qwen 3.6 27B Dense, Q4 (single RTX 3090) the undisputed king of the 24GB tier. one shot a playable game, around 41 tok/s, fits with context headroom to spare. one 24GB card, this is your answer. > 3. NVIDIA Nemotron 3 Nano Omni, 30B-A3B (DGX Spark) the best multimodal i've tested for video classification work. vision in, runs clean on the Spark. the rest, ran them, they hold up fine: on the Spark: DeepSeek V4 Flash 158B, GLM 4.7 Flash, GLM 4.5 Air REAP 82B-A12B, Gemma 4 26B-A4B, Qwen3-VL 235B-A22B, Qwen3 Coder 30B-A3B, Qwen3 30B-A3B, Carnice 35B-A3B. on consumer GPUs: Kimi K2.5 1T, Qwen3-Coder-Next 80B, Hermes 4.3 36B, Qwen 3.5 27B Dense. single 3090 to a 128GB Spark, that's the range. the three up top are the ones worth your hardware today.

@sudoingX its better then even sex

buy a gpu. 3090, 4090, dgx spark, whatever fits your budget. tier doesn't matter. running your first local model does. the moment your first prompt lands with no api between you and the model, your brain rewires. that single moment is worth more than every take you'll ever read on a timeline.

@teortaxesTex brazil runs on álcool

wtf is Brazil doing? All their cars work on ethanol?!

@SantosF80731369 @laurojardim acharam que ia desmobilizar o gado e o gado ta como? TOMANDO YPE KKKKKKKKKK é mta unidade partidária. imagina no tempo do césio

@laurojardim Esses "aliados" tomaram foi no cu, acharam que esse Papinho iria desmobilizar a militância e no fim do dia o saldo é positivo para Flávio, agora q meta processo em geral que fez ilações.

Aliados de Flávio Bolsonaro estão desnorteados dlvr.it/TSWwJ0

@a_lost_future @izakaminska now imagine being me having to read this shit even though i dont follow her just because she pays this stuff and gets her tweets to be shadow promoted. twitter is so depressing. *sigh* americans are so dumb

@izakaminska are you an anti-vaxxer/‘just asking questions’ type? If so, will just make an unfollow decision easy

Whoa. 🫣 This is incredible. Stephen Hoge, president of Moderna, gave an interview on April 17 to the "why should I trust you?" podcast in which he referred to the spike protein — famously in all the Covid vaccines — as err... "garbage". The exact quote is as follows and comes after he talks about the newest version of Moderna's Covid vaccine being superior because it "doesn't have all the spike protein stuff that we think was creating some of the problem". 👇 "And even now, there's some real world data that we've just put up on a preprint that's getting published showing that it's actually superior in the real world. And that innovation, we think is great, it's a lower dose. It doesn't have all the other spike protein garbage. It's more targeted" podcasts.apple.com/us/podcast/spe…

@gzhdigital que baita conteúdo! É uma profundidade jornalística sem rival no hemisfério

Thor e Olin: quem são os filhos de Eike Batista chamados de geração fru-fru pelo pai gauchazh.clicrbs.com.br/donna/gente/no…

@sudoingX yeah, and how much usage on zdr provider of the small model you are using? infinite?

fun fact: 4 years of $20/mo to chatgpt = the price of a used 3090. one leaves you with nothing. the other leaves you with a 3090.

I am sharing a set of slides from my lecture on agentic AI tools for Applied Economists at @penn_state. Materials are available here: github.com/ballesterogh/A… This builds on this thread and great work by Paul Goldsmith-Pinkham, @aniketapanjwani, and @VelikovMihail

Sci-Hub is an evil website that pirated 85M+ research papers and made them freely available And now they've added AI to their database to make Sci-Bot. It answers your questions using latest, full-text articles. But DO NOT use it. We should all try to make billion-dollar academic publishers richer. I'm putting the link below so you know how to avoid it.

The most powerful real-time visual tool in creative coding also has the steepest learning curve Now your Hermes agent can just run TouchDesigner for you. Video credit: made by @macbethAI, a talented AI artist and avid Hermes user, with the TouchDesigner skill

@batistaprimo @laurojardim realmente a desenformação é um problema grave

@laurojardim L. Jardim não existe polarização entre ricos e pobres, entres lulistas Bolsonaristas, o q existe é a desenformação d globo, q atua 24 horas c fatos inverídicos contra Gov. Lula, tenta sem sucesso, desestabilizar o cenário politico favorável à Lula, não consegue Lula Tetra 1° T...

Eleitores mais ricos são os mais polarizados entre lulismo e bolsonarismo, aponta pesquisa dlvr.it/TSFTHg

evolutionary pressures will create human vat brain lurkers that only wake up at off peak compute hours to exercise their will like the undersea life that eats marine snow

@sudoingX watabout context length? aren't you hitting your head on the ceiling multiple times?

polishing the autonomous build video next. honestly never gets old watching a 27 billion parameter model on a single 3090 crank through a problem, find its own bugs, patch them, re run tests, until everything lands green. dropping it shortly.

pay attention anon. this is what local ai actually feels like in 2026. qwen 3.6 27b dense just knocked down the second test in my single file agentic benchmark series. on 1x 3090. mandelbrot fractal explorer with zoom, pan, three palettes, smooth coloring, responsive layout, 800×600 canvas. autonomously built it from one prompt at around 30-40 tok/s on a single rtx 3090. this is no human intervention. the model wrote the html, wrote tests.js with 10 verifiable tests, served the page, opened it in its own browser, ran the tests, found failures, traced the math, patched the code, re ran, landed all 10 green on its own. second single file test in a row this model has cleared on the same hardware. that is a 27 billion parameter model on consumer hardware doing the kind of engineering work builders are paying twenty to a hundred dollars a month to outsource to closed apis. the question is not whether models can code anymore. the question is whether you have stopped paying api bills for the work your gpu can already handle locally. four screenshots from the live build.

@blogdosakamoto e como fica o patrão???

Fim da 6x1: Planalto rejeita subsídios e pagamento por hora como troca… - Veja mais em noticias.uol.com.br/colunas/leonar…

Understanding regression models is essential in data science. In 4 minutes, I'll demolish your confusion. Let's go:

Looks like RAM’s back on the menu boys!