I would like to know if it's possible to use Phi-3-Vision (exl2) with ExLlamaV2. I've seen that recently was some progress with multimodal tokens here but I have no idea how to use it or where it is documented (or if it's even in a working state). Phi-3-Vision seems quite capable from their online demo so I think it would be super cool to have some kind of support for it.
Thanks again for this amazing library and kind regards,
Timon
Hello @turboderp
I would like to know if it's possible to use Phi-3-Vision (exl2) with ExLlamaV2. I've seen that recently was some progress with multimodal tokens here but I have no idea how to use it or where it is documented (or if it's even in a working state). Phi-3-Vision seems quite capable from their online demo so I think it would be super cool to have some kind of support for it.
Thanks again for this amazing library and kind regards, Timon