Dave Batista 🐺

@manodeyvin Cadê o boné?!

quando a matéria não é patrocinada

@tylernotfound @sherwinwu this is more cool chatgpt.com/agents/studio/…

@isDaveBatista @sherwinwu More cool/informative 404 page otw 🫡

Really proud of some of the product details that went into ChatGPT Workspace Agents – like this fun little interaction on the "build a new agent" page. Has @tylernotfound's fingerprints all over it chatgpt.com/agents/studio/…

@luksamuk kkkkkkkkkkk

Por mim que enfiem o Kimi no Opus da Anthropic

@sama Interessante você dizer isso @sama , mas seria maravilhoso aumentar 1.7x o limite do codex, ai sim seria muito IA

We want you to have a lot of AI!

@devcelio Eles querem remover o claude code como produto apos o vazamento do código

vi muita gente postando e aparentemente a anthropic tirou o claude code do plano de R$100/$20 quem tem assinatura pode confirmar essa info?

@robinebers Você falou bem "permite testar", pois acaba super rápido, pelo menos no meu caso de uso

OpenCode Go is genuinely the best subscription EVERYONE should have - it’s just $10/m and lets you test the latest open source models for extremely cheap

@realsigridjin funciona certinho ou é só Hype Sigrid?

i am kimi poor

@abhitwt Tokens

i am a Vibe Coder, scare me with one word

@thsottiaux aumenta 1.7x o limite e está tudo certo no plano Pro, fica perfeito, Codex é muito bom, eu gosto do modo Fast. @romainhuet

Hello builders. What are we getting wrong with Codex, what can we improve?

Tim Cook is a legend. I am very thankful for everything he has done and I am very thankful for Apple.

@_falsi1ke Olá!

No account should be under 1K followers. Say hello, I will boost you

@NanouuSymeon Não uso nenhuma dessas, eu uso Codex, Claude Code e Droid

🤖 AI Tools Developers Are Using 🤖 GitHub Copilot 🤖 Codeium 🤖 Cursor AI 🤖 Phind 🤖 Perplexity AI 🤖 Tabnine 🤖 Replit Ghostwriter 🤖 Sourcegraph Cody

@XFreeze tem uma pegadinha ai, as tools são cobradas separadamente

Grok 4.1 Fast is like a token glitch for OpenClaw 🦞 It's an extremely fast, decently capable model that works GREAT with tool calls If you want to save money, this is the one. You're probably not going to use it for heavy coding, but for deep research, general tasks and other things, it hits the perfect cost-to-performance sweet spot while offering everything you need This model is exceptional at search. It performs close to GPT-5.4 (the very next model on the search Arena currently) The pricing is dirt cheap: - $0.20/M input | $0.50/M output - 2M Context window - Blended ~$0.26 per million tokens → 15–25× cheaper than OpenAI GPT or Claude Sonnet ($5–$15+) Plus, Native Web Search + 𝕏 Search tools are just $5 per 1,000 calls (literally pennies per query) With a 2M context window, it's an agentic beast for research, support, and deep dives You don't need to pay 20× more when Grok 4.1 Fast offers near-instant speed + real capability at a fraction of the cost

🚀 Novidades da @OpenAI! O Codex ganhou superpoderes: ✅ Usa apps no Mac ✅ Conecta mais ferramentas ✅ Cria imagens ✅ Aprende com ações passadas ✅ Lembra seu estilo de trabalho ✅ Executa tarefas repetíveis Codex: uma ferramenta para (quase) tudo! 🤖🔥 #OpenAI #Codex #IA

@ThePrimeagen Hoje ainda é bem improvável.

I cannot wait for gpt 5.5 here in ~30 minutes

@MatthewBerman O @bridgemindai é uma fonte confiavel, ele disse que iria ser lançado hoje e foi.

Unexpected drop

@claudeai @bridgemindai

Introducing Claude Opus 4.7, our most capable Opus model yet. It handles long-running tasks with more rigor, follows instructions more precisely, and verifies its own outputs before reporting back. You can hand off your hardest work with less supervision.

Saudades das lives de Hackathon do @RayFernando1337

@FirstSquawk Está tudo ficando caro ultimamente, pode esperar o preço subir uns 20 a 30% até final do ano

ANTHROPIC SHIFTS TO USAGE-BASED BILLING, INCREASING COSTS FOR HEAVY USERS - TIF