Z80-μLM

AI

tiny language model

SHOW HN2025

514points
122comments
#67of 3,837 in 2025
Top 2%of 2025 by points

How HN took it

MixedNo clear lean

21 positive, 19 neutral or asking, 1 mixed, 4 critical, out of 45 comments about the project. Thread score +0.37.

Idea: 11 for, 0 againstPerformance: 2 for, 3 againstDesign: 0 for, 1 against

The thread, comment by comment

Each row is a top-level comment and its replies, in HN's order. Select a cell to open that comment on HN.

positive neutral or mixed critical about another commenter or something else the maker

What HN said

What people liked

  • Impressive constraints and fun demo

    From 3 comments
    • This is super cool. Would love to see a Z80 simulator set up with these examples to play with!
      Zee2 on HN
    • An LLM in a .com file? Haha made my day
      alfiedotwtf on HN
    • Between this and RAM prices Zilog stock must be up! Awesome hack. Now apply the same principles to a laptop and take a megabyte or so, see what that does.
      jacquesm on HN
  • Nostalgic appeal of running it on retro hardware

    From 3 comments
    • Imagine, this working on a Gameboy, in those days. Would've sounded like magic
      codetiger on HN
    • This is excellent. Thing I’d like to do if I had time: get it running on a 48K Spectrum. 10 year old me would have found that absolutely magical back in the 1980s.
      bartread on HN
    • Awesome! Anyone for a port to the MSX? A web version would also be cool.
      DrNosferatu on HN

Objections

  • Would be very slow on real hardware

    From 1 comment
    • It runs, but it would be very slow on actual hardware. I tried on a cycle-accurate emulator of a TRS-80 Model I with Omikron CP/M mapper. Most Z-80 machines of the time were…
      jrdres on HN
  • Hard to get correct answers in the 20 Questions game

    From 1 comment
    • I've been playing the Z80-μLM demos in your CP/M emulator. Works great! However, I have yet to guess a correct answer in GUESS.COM! I'm not sure if I'm just not asking the right…
      jaak on HN

Requests

  • Port to MSX, Spectrum, or other retro machines

    From 3 comments
    • This is excellent. Thing I’d like to do if I had time: get it running on a 48K Spectrum. 10 year old me would have found that absolutely magical back in the 1980s.
      bartread on HN
    • Awesome! Anyone for a port to the MSX? A web version would also be cool.
      DrNosferatu on HN
    • Luckily I have a very large amount of MSX computers, zx, amstrad cpc etc and even one multiprocessor z80 cp/m machine for the real power. Wonder how gnarly this is going to…
      anonzzzies on HN
  • Web version

    From 1 comment
    • Awesome! Anyone for a port to the MSX? A web version would also be cool.
      DrNosferatu on HN

Questions

  • How was quantization-aware training done?

    From 1 comment
    • Did you train the model with quantization awareness? How?
      lostmsu on HN
  • Was sparse weights considered?

    From 1 comment
    • Very cool. Did you consider using sparse weights?
      Y_Y on HN
  • Tokens per second?

    From 1 comment

Alternatives named

Category, keywords, summaries and labels on this page were made by AI (Thread Reader v2) and not checked by hand. Each point links to the comments it came from.

Add a badge to your READMEShown on HackerShow
[![Shown on HackerShow](https://hackershow.org/badge.svg)](https://hackershow.org/show/46417815)