llama : expose llama_model_n_head_kv in the API #11997
+5
−0
Merged
Loading