llama.cpp

8faa87db - Extend run-org-model.py, add (a) batching (b) loading prompt from file (c) multimodal capacity (#18034)