nryoo

Were the withdrawn ones actually Lean-checked or not? seems like that matters

Picking lunch menu..?

good job :) kind of hard for me lol

One of their three identical accounts had ~20% lower limits and the explanation was an "extremely tiny" A/B test. Makes the "5x" kind of a joke. I'd honestly take a slightly worse plan if it just published the actual token limits.

Didn't expect to see CATL among the biggest backers. What does the battery company want with DeepSeek? Maybe they chase a chance to sell batteries to the datacenters?

Oh! sorry for that, I had a mistake :( I can't edit url, so I have to edit title for this submit and submit another contents. sorry again and thank you comment

The root problem is that coding agents treat tool output as trusted context.

Good idea. Worth considering though, philosophy isn't really about right answers, it's about learning to question. Bit worried about an LLM tends to hand kids a tidy conclusion, which is kind of the opposite of what you want.

The FFmpeg/Python step is neat. They avoid stopping on every frame, capture continuous video instead, then split it back out and pick the sharpest frame from each group.

From Atom Computing. The toric code needs non-local connections, which plays to neutral atoms' all-to-all connectivity over fixed-topology superconducting qubits.

Text summary since it's a video: it's a model abstraction layer. You write against a LanguageModelSession, import a Swift package, initialize the model you want, and pass it to the session. Apple ships SystemLanguageModel (on-device), PrivateCloudComputeLanguageModel, and open-sourced CoreAILanguageModel and MLXLanguageModel for local models; Anthropic and Google are publishing their own packages.

Microsoft dropped the legal threat, but it also dropped the phrase "responsible disclosure." The new statement says coordinated vulnerability disclosure instead. That's the term Microsoft itself switched to back in 2010, specifically so researchers who go public wouldn't be painted as irresponsible. Katie Moussouris, who helped make that switch, said invoking "responsible" this time was "the first strike in my book."

Nice writeup. One thing worth adding to the limitations: without vorticity confinement, the Gauss-Seidel projection step quietly dissipates the small-scale curl that makes smoke look like smoke.

The 2001 Fedkiw/Stam/Jensen "Visual Simulation of Smoke" paper added it back as a correction force for exactly this reason. At N=16 it doesn't matter much because the grid itself can't represent fine vortices, but the moment you crank N up the missing confinement becomes visible.

How big is the lifetime holes thing in practice? On loops the contiguous-interval model spills way more than it should. Wondering if that alone explains most of the YJIT gap.

So what's the good point?

So the state of AI in 2026: ChatGPT DDoS-lite, Claude the polite one that actually reads the rules, Perplexity maybe shows up, and Google was already in your house.

$180/month to control your lights and music. A Raspberry Pi + Home Assistant does this for $0/month and doesn't exfiltrate your home network topology to a third-party API. The value proposition only makes sense if your time is worth more than your privacy.

The real metric is: does it solve my problem, and is the maintainer still responding to issues? Everything else is just noise.

Open source helped me more than cold outreach. Shipping something small and useful gave potential clients something concrete to evaluate instead of just a resume. The conversations that followed were much warmer.

The EU angle is interesting but the more immediate concern is the opt-in/opt-out framing. Google historically defaults these features to on and buries the toggle — did they actually make face data access opt-in from the start, or is this another "we'll notify you and assume consent" situation? The GDPR enforcement gap between announcement and actual compliance has been wide enough to drive a truck through.