joke-finetome-model-gguf-phi4-20260112-081758/inference/llama_cli_examples.md

Local inference (llama.cpp)

llama-cli -hf {REPO_ID}:q8_0 -cnv --chat-template phi4

Server (OpenAI-compatible)

llama-server -hf {REPO_ID}:q8_0
# /v1/chat/completions will be available (OpenAI-compatible)