I am fascinated by this and similar research (RotorQuant, etc). It seem by next year we will be able to run this year's largest models on last year's hardware. :)

Maybe we won't need as many data centers and as much power as we thought. Maybe we can run more powerful models locally.

TurboQuant IS restricted EDEN quantization (NeurIPS 21, ICML 22). Missing the optimal scale derivations, which makes TurboQuant variant considerbly LESS accurate than those works. We show this throughfully in a new note in https://arxiv.org/abs/2604.18555

We were the first to introduce post-rotation distribution aware quantization in 21, which was LATER implemented in many fields including federated learning, vector retrieval, databases and KV-cache.

It would be nice to get some credit of this. And it is certainly baffling to see the name "TurboQuant" repeated in this context, considering the many works from 21 onwards.

This blog post basically goes you through EDEN quantization, but then ends with showing a less than optimal MSE-minimizing version and an unbiasing trick that is less optimal by a full bit in many cases.