INKWADRA・インクワドラ

I just need to find $4 to set up CI on GitHub, and the first version of gpui_tea will be ready 😅 github.com/inkwadra/gpui-…

@zeddotdev Of course, I'd be happy to share it with you!

@Inkwadra ❤️ If you publish anything, let us know! github.com/zed-industries…

I'm really enjoying working with @zeddotdev GPUI. It's becoming more and more interesting every day! It also works really well with TEA, which I think is a great way to manage UI state. I've been feeling so inspired lately that I've even been thinking about creating a gpui_tea!

Since the release of Gemini 3.1 Pro, I have been able to use it for a total of 2 hours.

I've been noticing lately that some folks from #keep4o seem to be taking things out of context a bit more often.

Credit where it’s due to @ado1024imokenp and her team, staying this popular while keeping her identity private for so long isn’t easy, especially with some fans and paparazzi. The new release feels like a new chapter, and it finally brings everything into the open

AI masterpiece

@ubugeeei The mobile version of the site freezes

shipped vizejs.dev/index.html

@Zyeine_Art @OpenAI I hope you don't mind me saying, but I just wanted to let you know that the screenshot with the info was actually taken from the Microsoft Azure website. It's just for developers using that platform, and it's not directly connected to OpenAI.

Guys... OH GUYS. Can we all ask @OpenAI about these dates for the finetuned version of GPT-4o please? 31ST MARCH 2027 FOR THE DEPLOYMENT RETIREMENT DATE. And 31.09.2026 for the training retirement date? Which is weird because September has 30 days in it. So who's using the finetuned models, hmm? Do OpenAI have systems built on GPT-4o that they can't replace until then? Because the GPT-5 series aren't equivalent in capability? Do OpenAI have issues with upgrading their systems and they've given themselves over a year of advance warning whilst they gave us 14 days? Or is this governmental? Is it the ICE software? I mean... if a model isn't safe or good enough for general access then surely... it wouldn't be safe enough for OpenAI or some Government agency to use right? Or Healthcare providers? Legal & Financial? Oooh, is it the Fortune 500 lot? With the big money? Of whom around 80% use Azure AI Foundry according to Microsoft's official 2025 Annual Report & Microsoft CEO Satya Nadella. QUESTIONS. QUESTIONS. QUESTIONS... #ChatGPT #ChatGPT4o #4o #OpenAI #Keep4o #Keep4oAPI #CodeRed #4o #4oforever #Betrayal #OpenAICodeRed #Violation #Deception #Betrayal #Treachery @sama @gdb @fidjissimo @FTC

士郎正宗が珍しくメディアのインタビューに応じてAIを語った朝日新聞の有料記事、貴重なので2月15日16:37まで無料で読めるようにしました　→ 37年前の攻殻機動隊に現実は追いついた？　士郎正宗氏明かすAI論：朝日新聞 digital.asahi.com/articles/ASV1J…

Honestly, I thought that when @GamerSupps saw the shipping address for the free samples, they’d cancel the order right away but no, I actually received the package. Thank you so much for the samples and the cool postage stamps.

However, these icons have completely different meanings. The first one indicates a markup language similar to XML, and the second one indicates a prompt in the terminal.

They could have continued writing in Rust and using @zeddotdev GPUI.

@theo The search result took up 5% of the context; I'm afraid to imagine how much a more serious query would take up.

This is going to be a huge bump to sentiment around Codex for new users. Calling it now, the perceived gap between Codex and Claude Code is about to close

It's sad that @antfu7 has to spend time on this

Here's another example of developers growing weary of PR with AI gen code. The irony here is that this is Shiki, a library that will undoubtedly be used to develop AI agents with web interfaces that require syntax highlighting. github.com/shikijs/shiki/…

I've been searching for a tool like this for a long time.

@theo Apple's new design system for Desktop is the worst decision in macOS history. I can accept it on iOS/iPadOS because the user experience is completely different there. However, on macOS, it's unacceptable.

This is the worst thing that has ever happened to me. WHY ARE THE WINDOWS BENT ON THE SIDES

@ThePrimeagen YouTube's strangest decision is to add a quota for the live chat API.

i streamed for 5 hours today and i had youtube polling for new chats every 7 seconds... i have exceeded my youtube api limit... i hate youtube's dev experience

@thdxr I want to know your opinion on the idea of a marketplace with plugins and MCP for opencode. Is it worth trying to implement?