DigiNews

Tech Watch by Johan Denoyer

← Back to articles

Show HN: Shoehorn – Quantize any model down to run on your machine

Quality: 8/10 Relevance: 9/10

Summary

Shoehorn is a browser-enabled, open-source quantization tool that lets you fit language models to memory-constrained hardware by performing per-tensor mixed-precision optimization. It outputs GGUF v3 and is MIT-licensed, with a Rust-based quantizer and no llama.cpp code linked. The article includes installation steps, a live model-fit UI, and links to the GitHub repo and design history.

🚀 Service construit par Johan Denoyer