llama.cpp Released Version b11988 with Limite 1B Model Support
Release b11988 of llama.cpp introduces native architecture support for the Limite 1B model, updating both the GGUF parsing and conversion scripts. This update allows users to run the Limite 1B model locally using the library. This update expands the ecosystem of open-source models supported by llama.cpp, giving developers and enthusiasts more options for local AI inference. Keeping pace with newly released model architectures ensures that community hardware can immediately leverage the latest advancements. The pull request (#29433) was authored by Giuseppe Scrivano with assistance from Claude Opus 5.5, adding the 'Limite' architecture across the gguf-py, convert, and llama components.
## BACKGROUND
llama.cpp is an open-source software library that performs efficient inference on various large language models using the GGUF file format, acting as a core engine for many local AI tools. GGUF is a unified file format designed to store everything needed to run a large language model on community hardware.