by fadizg
0
Run offline LLMs with llama.cpp backend: real‑time token streaming, SHA‑256‑verified resumable downloads, chat templates, KV‑cache reuse across turns, and grammar‑constrained generation.
#sdk, #parsing, #llm, #kotlin-multiplatform-sample, #kotlin-coroutines, #file, #dependency-injection, #cryptography, #compose-multiplatform, #cache