We read every piece of feedback, and take your input very seriously.
To see all available qualifiers, see our documentation.
Llama 2 with grammar-based decoding (provided by llama.cpp).