Same! I use ollama a lot, but when I need to do real engineering with language models, I end up having to go back to llama.cpp because I need grammar-constraine

Same! I use ollama a lot, but when I need to do real engineering with language models, I end up having to go back to llama.cpp because I need grammar-constrained generation to get most models to behave reasonably. They just don't follow instructions well enough without it.