https://t.co/0Kqkwho5C4

Rewrote pokeget in Zig: zigdex. Single binary, embeds 1010+ sprites for offline use, 1/128 shiny odds, and it's stupid fast, optimized for shell scripts. Benchmark (hyperfine): ▓░░░░░░░░░ 0.75ms zigdex ▓▓░░░░░░░░ 1.28ms pokeget ▓▓▓▓▓▓▓▓░░ 3.95ms

Just created ClaudeMeter, a little macOS menu bar app that shows your @claudeai usage like a battery. Pulls straight from the desktop app or browser session, updates every minute. Tells you when the next reset is. All local. (Not for the Claude API usage, but feel free to create

Promise stays Find her

Shawww from #SilkSong

So @Cloudflare is making you pay to fix their bug. Here’s the situation: Modern frameworks like @nextjs App Router use the HTTP Vary header (defined in RFC 7231) to tell CDNs: “Cache HTML and RSC payloads separately, based on request headers.” Example: /page with no RSC

⚡️ OAK RESEARCH V2 IS OUT. The most powerful tool for crypto investors. 5 new features. 9 months of work. Premium membership. We are just getting started.

The countdown is on! Join us in 48 hours for a special announcement about Hollow Knight: Silksong! Premiering here: https://t.co/sPt1IzanDS

They have to deal with the other seven roles. Every day. Source: I’m one of the seven.

chat, is this good ux?

We just launched the Full Version of LlamaFeed, a major upgrade to our free dashboard! It includes: - Expanded data - Content filters - New sections - AI news recaps - Revamped mobile Available to Pro & Llama+ subscribers All features 👇

I think @DefiLlama just released a new UI 👀 It looks so friggin dope 🔥

🚨Aujourd’hui marque le lancement de OAK Research Une plateforme d’intelligence de marché crypto offrant des analyses, des rapports de recherche et des données sur les tendances et innovations clés de l’écosystème. Disponible dès maintenant : https://t.co/DN428JGLnr Sur OAK

Excited to share our latest product with you: LlamaFeed It's designed to be your go-to landing page for up-to-date info across crypto. LlamaFeed updates every minute and lets you organise and select the sections that matter most to you. Try it out at:

study

@NFTbark crypto was founded on anon culture and there is nothing better than not needing to know the location, gender, language, etc of the person I interact with on a daily basis and instead base it on the work you actually do. *mutes post*

778

Thank you, Gary, you can go now. #ETF #Bitcoin $BTC

Conclusion: MLX, with its focus on Apple silicon, opens up possibilities in machine learning, offering an efficient and visually engaging way to develop and understand models. Try it out! https://t.co/f7I3BtD2rN

Interpreting Results: A steep drop in loss indicates rapid learning. The final plot reveals how closely the model's predictions align with the actual data, demonstrating the effectiveness of the learning process. [7/8]

Step 4: Run the script. Navigate to the directory where you saved the .py file using "cd". Run the file by typing "python3 (name).py". The model learns the relationship between features and targets, minimizing loss over iterations, showcased in an evolving plot. [6/8]