llama.cpp adds llama serve: run Inkling 2 Mini GGUF locally with one command

simonguozirui · x · 2026-11-21

julienc (Georgi Gerganov, llama.cpp creator) demos llama serve -hf ggml-org/Inkling-2-Mini-GGUF:IQ4NL — pulling the IQ4NL quant of the newly released Inkling 2 Mini straight from Hugging Face and starting a local server in one line. The announcement image shows ggml introduced Inkling 2 Mini on Nov 20, alongside the new llama serve command that further simplifies local LLM serving.

Original post →

More from Infra

Infra channel →