Bonsai 2 27B: Near-Lossless Compression in a 9x Smaller Footprint

1 points by mpweiher


mpweiher

Is it just me or are local models improving much more quickly than the frontier models?

If so, that would be a very welcome development, with local/private inference becoming available to everyone.

DustyFuzzy

The fact that you get 98.2% of the benchmark performance after going from 16 bits per weight to 1.76 bits per weight really shows you how much space is wasted in these models. Wow.