17,000 tokens per second!! Read that again!
LLM is hard-wired directly into silicon. no HBM, no liquid cooling, just raw specialized hardware. 10x faster and 20x cheaper than a B200.
the "waiting …

Get 10 Free Meals and your first box ship free with code REALLIFELOREFM at https://bit.ly/4ertNwx

Watch 50 additional exclusive RealLifeLore videos on Nebula in Modern Conflicts: https://nebula.tv/modernconflicts

Please Subscribe: https://www.youtube.c…

A 5-20x faster experimental Homebrew alternative. Contribute to lucasgelfond/zerobrew development by creating an account on GitHub.

Configurable, smart and fast CSS, SCSS, Sass and Less formatter with dprint integration. - g-plane/malva

a fast, modern browser for the npm registry. Search, browse, and explore packages with a modern interface.

Better JIT for Postgres. Contribute to vladich/pg_jitter development by creating an account on GitHub.

Contribute to databricks/flashoptim development by creating an account on GitHub.

JOVIAL J73 compiler in C99. x86-64 and RISC-V 64 backends. - Zaneham/skyhawk

Talking to a voice AI LLM over ham radio (on UHF 420.69 megahertz, of course!)

(Note: cool experiment, but be careful: FCC regs require a licensed control operator to be present at the control point the entire time the LLM is operating.)

Explore machine learning models.