Shoehorn – Quantize any model down to run apps now

For most of the use cases people might want to trade speed/optimization for more runtime customizations which will inherently be slower than pre-baked SIMD regex and whatnot.

It is quite disappointing to see them not using the type of work I did, but it was just too risky to have a human proofread the first paragraph in your product launch? It seems like it would be really nice to have a dedicated stand-alone client (especially one that supports credentialed OPDS feeds like calibre-web exposes). Wonderfully written. I'm intrigued by the section about emotion. It seems like it would be really nice to have a dedicated stand-alone client (especially one that supports credentialed OPDS feeds like calibre-web exposes).