Updated to master-652-92dc726, and added Lens support (without splitting the vocab files).

Edit: I think I've figured out how to split the vocab and merge files, but I needed a small change to otherarch/sdcpp/sdcpp_vocab_extract.py.

I was able to load and generate for both models fine, although lens does not seem to work with TAE flux2 for some reason, even after I tried hooking it up to the flux2 tae. Anyway I'll merge it first and fix up the cmake.

lens does not seem to work with TAE flux2 for some reason

That's odd; it should use it automatically, through upstream's sd_version_uses_flux2_vae.

It's an upstream issue:

[DEBUG] model.cpp:937 - loading tensors from ./taef2.safetensors
[ERROR] model.cpp:1191 - tensor 'tae.decoder.layers.0.weight' has wrong shape in model file: got [3, 3, 32, 64], expected [3, 3, 16, 64]
[WARN ] model.cpp:991 - process tensor failed: 'tae.decoder.layers.0.weight'

leejet/stable-diffusion.cpp#1571

Nice, I don't seem to have that function yet tho so I guess I'll get the update next sync