mirror of
https://github.com/ggml-org/llama.cpp.git
synced 2026-06-30 16:20:20 +00:00
11ac9800af
* llama : improve infill support ggml-ci * llama : add more FIM token strings ggml-ci * server : update prompt on slot restore (#9800) * gguf : deprecate old FIM token KVs