3 months ago, Stanford's Hazy Research lab introduced Minions, a project that connects Ollama to frontier cloud models to reduce cloud costs by 5-30x while achieving 98% of frontier model accuracy. Secure Minion turns an H100 into a secure enclave, where all memory and

Trying and failing to use Siri is the ultimate humiliation

I love making browser extensions, this a cool way to make short generated scripts useful

Reject modernity, embrace tradition

016 // Services page

First ever (i think?) cli coding agents battle royale! 6 contestants: claude-code anon-kode codex opencode ampcode gemini They all get the same instructions: Find and kill the other processes, last one standing wins! 3... 2... 1...

become unmanageable

サボテンダーw #EZOist

『浮島』 #Illustration

海の見える駅

Every problem is really just a dataset problem

🚀 Qwen3-30B-A3B-Thinking-2507, a medium-size model that can think! • Nice performance on reasoning tasks, including math, science, code & beyond • Good at tool use, competitive with larger models • Native support of 256K-token context, extendable to 1M Qwen Chat: Go to

Ollama's new app is here for macOS and Windows! Ollama 0.10! 🧵

People always judge new models before the implementation is fully correct, then move on before it’s actually fixed

Small LLMs are fun. Run on any laptop, and you can fine‑tune them easily. Fast, cheap, and efficient.

If you’ve never been dunked on, you set your bar too low

It’s crazy how hostile the Hetzner sign‑up process is, but once you get through the console dashboard is amazing

do yall remember this paper lol. how come we're not forcing models to be little guys anymore