Theo Borges

@ThePrimeagen I like the Copilot subscription, though. The agent I don't.

Qwen is not shy of letting you know

basically everyone is telling us there's shortages in every component of deploying GPUs, even labor lot of nervousness and hoarding right now, some crazy stuff going on idk what things are going to look like in the next 6 months

you're probably underestimating how crazy things are

@pokerdev7 Faria por 8k USD, imagina 20k.

Um salário fixo de 20k USD por mês (corrigido pela inflação) pelo resto da vida PORÉM você não pode escrever nenhuma linha de código pelo resto da vida, APENAS orquestrar IA para escrever por você Você faria esse acordo?

@louisisnothere_ @sudo_goreng This $60 inference is not enough for me, but for light use it is great value. I burnt 20% of the week's limits in 1 hour of work.

@sudo_goreng considered subbing to this; but ended up on Opencode Go instead $60 of inference for $10 ($5 first month) - insane value

Or you can get 3x, forever, for the same model.

@ephraimduncan It's better to just use the Kimi K2.5 since it's basically the same model.

You can now use Composer 2 in OpenCode with your Cursor Subscription

@cgarciae88 @theo No. Codex is that good compared to opus. Opus is not the best coding model long ago

@theo this anti-opus agenda feels fake

Just let Opus go for over an hour on a new feature. When it was done, I asked how I can test it. 20 minutes later, it realized I can't test it because it did the whole thing entirely wrong. Idk how you guys use this model every day for real work 🙃

@ArealJEDI @MiniMax_AI @grok A LOT.

@MiniMax_AI @grok how much ram is needed to run this model locally

Introducing MiniMax-M2.7, our first model which deeply participated in its own evolution, with an 88% win-rate vs M2.5 - Production-Ready SWE: With SOTA performance in SWE-Pro (56.22%) and Terminal Bench 2 (57.0%), M2.7 reduced intervention-to-recovery time for online incidents to 3-min on certain occasions. - Advanced Agentic Abilities: Trained for Agent Teams and tool search tool, with 97% skill adherence across 40+ complex skills. M2.7 is on par with Sonnet 4.6 in OpenClaw. - Professional Workspace: SOTA in professional knowledge, supports multi-turn, high-fidelity Office file editing. MiniMax Agent: agent.minimax.io API: platform.minimax.io Token Plan: platform.minimax.io/subscribe/toke…

@360_takiguchi Se você for colega dele, já dá umas ideias sobre outras profissões.

Aluno mais fraco da minha faculdade: #bolhadev #bolhatec #bolhasec

@juristr L3, and it has been more than enough.

What's your AI adoption level? (according to Steve Yegge)

@LukeberryPi O mais triste é que o AI Slop é melhor que 99%, tecnicamente falando, que quem fala "AI Slop"

a banalização do termo "ai slop" está absurda. e só vai piorar NVIDIA lança shader insano de upscaling em vídeo > "ai slop" Claude Code construindo sites inteiros e funcionais > "ai slop" LALAL.⁠ai cria o melhor stem splitter da história > "ai slop"

@jaomtts @JotaVictor97 @analise2025 Verdade, porém, em partes. Ele tem culpa sim.

@JotaVictor97 @analise2025 Da uma pesquisada q vc vai saber o porque do preço da gasolina alta, e pasmem n tem nada haver com o lula

🚨 URGENTE: Flávio Bolsonaro vai ENTERRAR o fim da escala 6x1 se for eleito! (Correio Brasiliense) Mesmo se passar no Congresso esse ano, ele vai retirar na Reforma Trabalhista que pretende fazer. O Bolsonarismo é inimigo do povo!

@ZattarRafael Qual é a má notícia?

Mano….

people keep saying AI replacing developers is just hype. meanwhile the creator of Node.js says: “This has been said a thousand times before, but allow me to add my own voice: the era of humans writing code is over. Disturbing for those of us who identify as SWEs, but no less true. That’s not to say SWEs don’t have work to do, but writing syntax directly is not it.” maybe we should listen.

Indeed. Sad truth.

@devcelio @akinncar Muita inveja dos manos coda fofo.

@akinncar Por isso eu postei no insta a timelapse :) instagram.com/reel/DV5v7_Djo…

ja pararam para pensar que ele pode só ter ligado o cronometro, ido dormir, acordado, e tirado a foto?

@theo Is cmux only for Mac? No Linux? I have both but use linux more. R.I.P.

So I've been using a new terminal...

@pokerdev7 Nenhum lugar tem uma bolha de IA tão grande quanto o X. Porém, o problema não foi o YouTube em si, mas sim o @manodeyvin, que entrou nessa de usar agentes mais tardiamente.

claude code chegou só agora no youtube?

@ZaiforStartups @grok how much smaller GLM5 Turbo is than the GLM5 model.

AI hardware is starting to appear. But running reliable agents is a different challenge. Tool use. Long tasks. Multi-step workflows. GLM-5-Turbo + OpenClaw now run inside the Mechrevo AI box bringing native agent runtimes to devices.