In the next version of Bun We made `postMessage(string)` 500x faster

🤨🥸

This is huge!

🔥

We’ve found and resolved two issues that were affecting quality in some Claude responses. We are continuing to monitor for any ongoing quality issues. We're grateful to the detailed community reports that helped us identify and isolate these bugs.

Kind of crazy that Claude models have had crazy quality issues nonstop for like 3 weeks now and...nobody's talking about it?

reminder to cancel your Claude subscription. why? > LYING and DENYING their models DEGRADED QUALITY, even their DevRel tried to gaslight me it in previous tweets of mine (he's nowhere to be seen now though lol) > 1.58-bit quantized models during daytime > plus not getting opus

Claude Code shipped three cool inline commands: - think (low) - think hard (medium) - ultrathink (max)

Claude Code /context command:

$100,000 fee per year for H-1B visas, Is this a joke?!

Did you notice Anthropic is the only company making efforts to regulate AI with law? This is so scary and is really bad. Dominance that comes in this way can easily harm the AI ecosystem.

Google started to release specialized series of Gemini models. Huge step forward! 🎉

I cancelled my Claude Code subscription.

My claude code session reached its 5 hour limit right now at 01:17 am. The error message says: 5-hour limit reached . resets 5am So, this means I only used 17 minutes of Opus, and after that the limit reached and will reset after around 4 hours. Anthropic Claude

Sharing our second Connectionism research post on Modular Manifolds, a mathematical approach to refining training at each layer of the neural network

Anthropic Claude I have MAX subscription, and just in 4 minutes of Opus 4.1 usage, the “Approaching Opus usage limit” warning is appeared. I am requesting an answer from you here since there is no customer service there. Why 4 minutes usage should trigger that warning?

Finally, Claude Sonnet 4.5 🎉