The performance numbers do not show time performance versus accuracy perforamnce. Binary embeddings clearly have a time advantage, at the expense of accuracy. Reporting/benchmarking in this domain requires (at least) two dimensions.
I FT'd some siglip models (see here: https://huggingface.co/carsonpoole/binary-siglip-text) that should be amenable to binary quantization so I'm going to tomorrow get an inference server running with that and then hopefully do the typical MTEB benchmarks for embeddings
[0] https://github.com/facebookresearch/faiss/wiki/Binary-indexe...