# llama.cpp Compatibility

Vision encoder based on Mistral Small model, supports image-text generation tasks, compatible with llama.cpp framework

The Llamacpp imatrix quantized version of Qwen3-0.6B provided by the Qwen team, quantized using llama.cpp, supports running in LM Studio or projects based on llama.cpp.

This is the quantized version of the pixtral-12b model, quantized using llama.cpp, supporting image-text-to-text tasks.

This model is a GGUF-format version converted from BAAI/bge-reranker-v2-m3, primarily designed for text ranking tasks with multilingual support.