rpdaiml
Re:Show HN: Bonsai 1.7B ternary model at 442T/s on M4 Max
on Monday May 04, 2026 @03:57PM (#48014117)
Nice work, that throughput is wild.
Re:Show HN: Muesli – If Granola and Wisprflow had an open source on de...
on Monday May 04, 2026 @03:56PM (#48014097)
How are you handling the on device speech pipeline, especially around model size, latency, and accuracy tradeoffs on consumer hardware?
Re:Show HN: I built a tiny LLM to demystify how language models work
on Monday April 06, 2026 @10:44AM (#47661611)
This is a nice idea. A tiny implementation can be way more useful for learning than yet another wrapper around a big model, especially if it keeps the training loop and inference path small enough to read end to end.
Re:Show HN: 1-Bit Bonsai, the First Commercially Viable 1-Bit LLMs
on Thursday April 02, 2026 @11:10AM (#47615521)
curious and wondering its porting and usage on edge devices?