Wow, the operation cheapseek has ended?? am i the only one experiencing issues with @deepseek_ai via @opencode go?

Great idea but v poor implementation.... Just look at the x-axis, labels and colored areas...

@morganlinton @VulcanBench

Seems like we have a new token saver from @jetbrains team — the benjamin-plus hook. gonna try it today, benchmarks look promising. One should expect 10-18% cost savings.

am i the only one or the grok is down in Europe? was fine an hour ago and now I even cant start Grok Build

I've been using non-VS Code based IDEs more and more lately. But I found that I became so used to the @system76 Pop Dark theme created by @ArtisanByteCrafter So I have ported it for @zeddotdev and the Pi. I cannot help myself, I love these warm colours, especially in Zed.

Hmm, the shared link seems to be corrupted. Anyways, just check this table recognition. All I've done is uploaded the PDF paper to the ocr.z.ai

From my experience the way LLM converts PDF to md matters most. By default it probably uses very quick parsers, which can lead to information loss. Here is an example of using the marker lib to convert the Elseveir's paper. Left is quick, right is slow with postprocessing

I know I'm biased, but maaaan, Formalist and Air work so well together in the latest Positron update. Dear #rstats, you should give Positron a try. 💨 Air — posit-dev.github.io/air/ 🎩 Formalist — github.com/atsyplenkov/forma…

Don’t get me wrong, I am not trying to discover hot water. Just sharing my observations. Btw, A v nice intro to @cursor_ai features by @rasmic youtu.be/gqUQbjsYZLQ?si=mU2f…