Llama.cpp Server cache_prompt
Adds the cache_prompt option for llama-server completions, enabling users to reuse the KV cache when supported, improving completion speed in long conversations when using llama.cpp server. See: https://github.com/ggerganov/llama.cpp/blob/master/examples/server/README.md#post-completion-given-a-prompt-it-returns-the-predicted-completion
README
No README available