feat: add Q4_K_M + Q8_0 + BF16 gguf for Ollama / llama.cpp consumers 0dedfe4 Snider Virgil commited on 4 days ago