Z80-μLM: A 2-Bit Quantized Language Model for Vintage Z80 Processors
By
quesomaster9000
If you only eat one bagel today, this is the bagel.
Summary
Z80-μLM is a 2-bit quantized language model designed to run on vintage 8-bit Z80 processors with only 64KB of RAM. The project explores how small a conversational AI can be while maintaining personality, resulting in a 40KB .COM binary that can run on 1976-era 4MHz hardware. It allows training conversational models in Python and exporting them as CP/M binaries for retrocomputing enthusiasts to chat with their vintage computers.
Key quotes
· 4 pulledZ80-μLM is a 'conversational AI' that generates short character-by-character sequences, with quantization-aware training (QAT) to run on a Z80 processor with 64kb of ram.
The root behind this project was the question: how small can we go while still having personality, and can it be trained or fine-tuned easily? With easy self-hosted distribution?
The answer is Yes! And a 40kb .com binary (including inference, weights & a chat-style UI) running on a 4MHz processor from 1976.
It won't pass the Turing test, but it might make you smile at the green screen.
You might also wanna read
Reflections on DwarfStar 4's rapid rise in local AI inference
The author reflects on the unexpected popularity of DwarfStar 4 (DS4), a local AI inference project. They attribute its success to the conve
antirez.com·1d agoReflections on DwarfStar 4's rapid rise in local AI inference
The author reflects on the unexpected popularity of DwarfStar 4 (DS4), a local AI inference project. They attribute its success to the conve
antirez.com·1d ago
Building a Personal AI Agent with Markdown-Based Skills and Local Models
The article describes a personal AI agent built on Pi that manages the author's inbox, calendar, deal pipeline, blog publishing, and researc
tomtunguz.com·2d ago
StepFun Releases Step 3.5 Flash: 196B Sparse MoE Model for OpenClaw Agents
StepFun has released Step 3.5 Flash, a 196B sparse Mixture of Experts (MoE) model that activates only 11B parameters per token for high effi
Product Hunt·2d ago
Anthropic Releases Claude Opus 4.7 AI Model with 1M Context Window and Enhanced Coding Capabilities
Anthropic announces Claude Opus 4.7, their latest AI model featuring a hybrid reasoning architecture with a 1 million token context window.
anthropic.com·3d agoAnthropic Releases Claude Opus 4.7 AI Model with 1M Context Window and Enhanced Coding Capabilities
Anthropic announces Claude Opus 4.7, their latest AI model featuring a hybrid reasoning architecture with a 1 million token context window.
anthropic.com·3d ago