Shoehorn

Quantizes a BF16 GGUF model to exactly fit your Mac's VRAM, then runs it with llama.cpp.

Desktopfreeglobal

Shoehorn quantizes a BF16 GGUF language model to exactly fit the memory an Apple Silicon Mac actually has, solving a per-tensor mixed-precision assignment guided by an importance matrix instead of using fixed presets. One command downloads, quantizes, and serves the model with llama.cpp. It is open source and macOS-only.

Website

Full match profile

Behind the summary, Matchbox keeps a richer profile of Shoehorn - the signals our matcher actually reads to decide when to surface it. It stays private; claim the listing to see and control it.

Problem & pain-point mapping
Who we surface it to (audience fit)
What it's a strong alternative to
Trust & credibility signals

Claim Shoehorn to unlock →

Try Matchbox with your own problem

Describe what is not working - we’ll show you whether Shoehorn (or something else) actually fits.

Try a problem