For llama.cpp, this would involve another handcrafted Siglip2 NN process. Interesting—they just figured out how to handle Siglip1. https://github.com/ggml-org/l

For llama.cpp, this would involve another handcrafted Siglip2 NN process. Interesting—they just figured out how to handle Siglip1. https://github.com/ggml-org/llama.cpp/pull/12344/commits/6313b97a1ed98c6e1cc9c8958e949fa7fe28b1cf#diff-d2bb0434579582f245f0130106c116dd642894fd4929d0c4775fefc4310cd729R642 https://github.com/ggml-org/llama.cpp/pull/12344/commits/631...