[Q] Do we support colqwen in this repo?

[yichuan-w]

I want to ask this question as I noticed that in the readme, it said you support Copali

[raphaelsty]

Hi @yichuan-w, thanks for asking! Yes, we do support colqwen 😀

The library itself is model-independent. We mentioned Copali in the README just as a concrete example of a model that fits the architecture, since it outputs k embeddings for a single input.

But you can use any model you want. As long as your model produces a set of k embedding vectors for a single input (whether it's text, an image, audio, etc.), you can use FastPlaid to index and search it :)

The model needs to have been trained using max-similarity function that's all, but this is the case for colqwen